From 409fb39717be9ab7becd9e8c01801a08c5bae318 Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Tue, 21 Jul 2026 13:53:30 -0400
Subject: [PATCH 01/29] refac

---
 backend/open_webui/models/chats.py            |  22 ++-
 backend/open_webui/routers/folders.py         |  22 ++-
 src/lib/apis/folders/index.ts                 |  39 ++++--
 src/lib/components/chat/Chat.svelte           |   2 +-
 .../chat/Placeholder/ChatList.svelte          | 129 ++++++++++++------
 .../chat/Placeholder/FolderPlaceholder.svelte |  72 +++++++---
 .../layout/Sidebar/RecursiveFolder.svelte     |  78 +++++++++--
 .../layout/Sidebar/SharedFolderItem.svelte    |  62 +++++++--
 .../(app)/folders/[folderId]/+page.svelte     |  46 +++++++
 9 files changed, 374 insertions(+), 98 deletions(-)
 create mode 100644 src/routes/(app)/folders/[folderId]/+page.svelte

diff --git a/backend/open_webui/models/chats.py b/backend/open_webui/models/chats.py
index c20c1db247..1d959f0bc6 100644
--- a/backend/open_webui/models/chats.py
+++ b/backend/open_webui/models/chats.py
@@ -1722,17 +1722,21 @@ class ChatTable:
         folder_id: str,
         skip: int = 0,
         limit: int = 60,
+        sort_by: str = 'updated_at',
+        sort_dir: str = 'desc',
         db: AsyncSession | None = None,
     ) -> list[dict]:
         """Get chats in a folder across ALL users. Returns dicts with user_id."""
         async with get_async_db_context(db) as session:
+            sort_column = Chat.title if sort_by == 'title' else Chat.updated_at
+            order_clause = sort_column.asc() if sort_dir == 'asc' else sort_column.desc()
             stmt = (
                 select(Chat.id, Chat.title, Chat.user_id, Chat.updated_at, Chat.created_at, Chat.last_read_at)
                 .filter_by(folder_id=folder_id)
                 .filter(or_(Chat.pinned == False, Chat.pinned == None))
                 .filter_by(archived=False)
                 .where(Chat.meta['internal'].as_boolean().is_not(True))
-                .order_by(Chat.updated_at.desc(), Chat.id)
+                .order_by(order_clause, Chat.id)
             )
 
             if skip:
@@ -1754,6 +1758,22 @@ class ChatTable:
                 for chat in all_chats
             ]
 
+    async def count_all_chats_by_folder_id(
+        self,
+        folder_id: str,
+        db: AsyncSession | None = None,
+    ) -> int:
+        async with get_async_db_context(db) as session:
+            stmt = (
+                select(func.count(Chat.id))
+                .filter_by(folder_id=folder_id)
+                .filter(or_(Chat.pinned == False, Chat.pinned == None))
+                .filter_by(archived=False)
+                .where(Chat.meta['internal'].as_boolean().is_not(True))
+            )
+            result = await session.execute(stmt)
+            return result.scalar_one()
+
     async def get_chats_by_folder_ids_and_user_id(
         self, folder_ids: list[str], user_id: str, db: AsyncSession | None = None
     ) -> list[ChatModel]:
diff --git a/backend/open_webui/routers/folders.py b/backend/open_webui/routers/folders.py
index a4ea4b8961..e3ebbfd96f 100644
--- a/backend/open_webui/routers/folders.py
+++ b/backend/open_webui/routers/folders.py
@@ -6,7 +6,7 @@ import uuid
 from pathlib import Path
 from typing import Optional
 
-from fastapi import APIRouter, Depends, File, HTTPException, Request, UploadFile, status
+from fastapi import APIRouter, Depends, File, HTTPException, Query, Request, UploadFile, status
 from fastapi.responses import FileResponse, StreamingResponse
 from open_webui.config import UPLOAD_DIR
 from open_webui.constants import ERROR_MESSAGES
@@ -482,6 +482,9 @@ async def update_folder_access_by_id(
 async def get_shared_folder_chats(
     request: Request,
     id: str,
+    page: int | None = Query(None, ge=1),
+    sort_by: str = Query('updated_at'),
+    sort_dir: str = Query('desc'),
     user=Depends(get_verified_user),
     db: AsyncSession = Depends(get_async_session),
 ):
@@ -505,7 +508,17 @@ async def get_shared_folder_chats(
             detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
         )
 
-    chats = await Chats.get_all_chats_by_folder_id(id, db=db)
+    limit = 10
+    skip = (page - 1) * limit if page is not None else 0
+    chats = await Chats.get_all_chats_by_folder_id(
+        id,
+        skip=skip,
+        limit=limit if page is not None else 60,
+        sort_by=sort_by,
+        sort_dir=sort_dir,
+        db=db,
+    )
+    total = await Chats.count_all_chats_by_folder_id(id, db=db) if page is not None else len(chats)
 
     # Resolve owner names for display (avatar URLs are constructed client-side)
     owner_cache: dict[str, str] = {}
@@ -516,10 +529,13 @@ async def get_shared_folder_chats(
             owner_cache[uid] = u.name if u else 'Unknown'
         chat['owner_name'] = owner_cache[uid]
 
-    return {
+    response = {
         'chats': [{**chat, 'readonly': chat['user_id'] != user.id} for chat in chats],
         'folder_permission': 'write' if has_write else 'read',
     }
+    if page is not None:
+        response.update({'total': total, 'has_more': skip + limit < total})
+    return response
 
 
 ############################
diff --git a/src/lib/apis/folders/index.ts b/src/lib/apis/folders/index.ts
index b79e588947..d82f591289 100644
--- a/src/lib/apis/folders/index.ts
+++ b/src/lib/apis/folders/index.ts
@@ -290,17 +290,40 @@ export const getSharedFolders = async (token: string) => {
 	return res;
 };
 
-export const getSharedFolderChats = async (token: string, folderId: string) => {
+export const getSharedFolderChats = async (
+	token: string,
+	folderId: string,
+	params: {
+		page?: number | null;
+		sortBy?: 'title' | 'updated_at';
+		sortDir?: 'asc' | 'desc';
+	} = {}
+) => {
 	let error = null;
 
-	const res = await fetch(`${WEBUI_API_BASE_URL}/folders/${folderId}/shared/chats`, {
-		method: 'GET',
-		headers: {
-			Accept: 'application/json',
-			'Content-Type': 'application/json',
-			authorization: `Bearer ${token}`
+	const searchParams = new URLSearchParams();
+	if (params.page !== undefined && params.page !== null) {
+		searchParams.append('page', `${params.page}`);
+	}
+	if (params.sortBy) {
+		searchParams.append('sort_by', params.sortBy);
+	}
+	if (params.sortDir) {
+		searchParams.append('sort_dir', params.sortDir);
+	}
+	const query = searchParams.toString();
+
+	const res = await fetch(
+		`${WEBUI_API_BASE_URL}/folders/${folderId}/shared/chats${query ? `?${query}` : ''}`,
+		{
+			method: 'GET',
+			headers: {
+				Accept: 'application/json',
+				'Content-Type': 'application/json',
+				authorization: `Bearer ${token}`
+			}
 		}
-	})
+	)
 		.then(async (res) => {
 			if (!res.ok) throw await res.json();
 			return res.json();
diff --git a/src/lib/components/chat/Chat.svelte b/src/lib/components/chat/Chat.svelte
index e77e296129..08bde0bc9f 100644
--- a/src/lib/components/chat/Chat.svelte
+++ b/src/lib/components/chat/Chat.svelte
@@ -1177,7 +1177,7 @@
 		}
 
 		const pageSubscribe = page.subscribe(async (p) => {
-			if (p.url.pathname === '/') {
+			if (p.url.pathname === '/' || p.url.pathname.startsWith('/folders/')) {
 				await tick();
 				initNewChat();
 			}
diff --git a/src/lib/components/chat/Placeholder/ChatList.svelte b/src/lib/components/chat/Placeholder/ChatList.svelte
index 64a9ed215d..77bd5ae4ab 100644
--- a/src/lib/components/chat/Placeholder/ChatList.svelte
+++ b/src/lib/components/chat/Placeholder/ChatList.svelte
@@ -1,27 +1,35 @@
 <script lang="ts">
-	import { getContext, onMount } from 'svelte';
-	const i18n = getContext('i18n');
+	import { getContext } from 'svelte';
+	import type { Writable } from 'svelte/store';
+
+	const i18n: Writable<any> = getContext('i18n');
 
 	import dayjs from 'dayjs';
 	import localizedFormat from 'dayjs/plugin/localizedFormat';
 	import { getTimeRange } from '$lib/utils';
 	import ChevronUp from '$lib/components/icons/ChevronUp.svelte';
 	import ChevronDown from '$lib/components/icons/ChevronDown.svelte';
-	import Loader from '$lib/components/common/Loader.svelte';
-	import Spinner from '$lib/components/common/Spinner.svelte';
+	import ChevronLeft from '$lib/components/icons/ChevronLeft.svelte';
+	import ChevronRight from '$lib/components/icons/ChevronRight.svelte';
 	import Tooltip from '$lib/components/common/Tooltip.svelte';
 
 	dayjs.extend(localizedFormat);
 
-	export let chats = [];
+	export let chats: any[] = [];
 
 	export let chatListLoading = false;
-	export let allChatsLoaded = false;
-
-	export let loadHandler: Function = null;
 	export let showOwnerInfo = false;
+	export let page = 1;
+	export let total = 0;
+	export let perPage = 10;
+	export let orderBy: 'title' | 'updated_at' = 'updated_at';
+	export let direction: 'asc' | 'desc' = 'desc';
+	export let onPageChange: Function = () => {};
+	export let onSort: Function = () => {};
 
-	let chatList = null;
+	let chatList: any[] | null = null;
+	let totalPages = 1;
+	let pages: (number | 'ellipsis')[] = [];
 
 	const init = async () => {
 		if (chats.length === 0) {
@@ -31,34 +39,43 @@
 				...chat,
 				time_range: getTimeRange(chat.updated_at)
 			}));
-
-			chatList.sort((a, b) => {
-				if (direction === 'asc') {
-					return a[orderBy] > b[orderBy] ? 1 : -1;
-				} else {
-					return a[orderBy] < b[orderBy] ? 1 : -1;
-				}
-			});
 		}
 	};
 
-	const setSortKey = (key) => {
-		if (orderBy === key) {
-			direction = direction === 'asc' ? 'desc' : 'asc';
-		} else {
-			orderBy = key;
-			direction = 'asc';
-		}
-
-		init();
+	const setSortKey = (key: 'title' | 'updated_at') => {
+		onSort(key);
 	};
 
-	let orderBy = 'updated_at';
-	let direction = 'desc'; // 'asc' or 'desc'
+	const buildPages = (currentPage: number, pageCount: number): (number | 'ellipsis')[] => {
+		if (pageCount <= 7) {
+			return Array.from({ length: pageCount }, (_, i) => i + 1);
+		}
+
+		const items: (number | 'ellipsis')[] = [1];
+		if (currentPage > 3) {
+			items.push('ellipsis');
+		}
+
+		const start = Math.max(2, currentPage - 1);
+		const end = Math.min(pageCount - 1, currentPage + 1);
+		for (let i = start; i <= end; i++) {
+			items.push(i);
+		}
+
+		if (currentPage < pageCount - 2) {
+			items.push('ellipsis');
+		}
+		items.push(pageCount);
+
+		return items;
+	};
 
 	$: if (chats) {
 		init();
 	}
+
+	$: totalPages = Math.max(1, Math.ceil(total / perPage));
+	$: pages = buildPages(page, totalPages);
 </script>
 
 {#if chatList}
@@ -153,7 +170,6 @@
 				class=" w-full flex justify-between items-center rounded-lg text-sm py-2 px-3 hover:bg-gray-50 dark:hover:bg-gray-850"
 				draggable="false"
 				href={`/c/${chat.id}`}
-				on:click={() => (show = false)}
 			>
 				<div class="text-ellipsis line-clamp-1 w-full sm:basis-3/5">
 					{chat?.title}
@@ -177,19 +193,48 @@
 			</a>
 		{/each}
 
-		{#if !allChatsLoaded && loadHandler}
-			<Loader
-				on:visible={(e) => {
-					if (!chatListLoading) {
-						loadHandler();
-					}
-				}}
-			>
-				<div class="w-full flex justify-center py-1 text-xs animate-pulse items-center gap-2">
-					<Spinner className=" size-4" />
-					<div class=" ">Loading...</div>
-				</div>
-			</Loader>
+		{#if totalPages > 1}
+			<div class="flex items-center justify-center gap-1.5 pt-2">
+				<button
+					class="inline-flex size-7 items-center justify-center rounded-lg text-xs font-medium text-gray-500 transition hover:bg-gray-50 hover:text-gray-700 disabled:cursor-not-allowed disabled:opacity-25 dark:text-gray-500 dark:hover:bg-gray-850 dark:hover:text-gray-300"
+					disabled={chatListLoading || page <= 1}
+					aria-label={$i18n.t('Previous page')}
+					on:click={() => onPageChange(page - 1)}
+				>
+					<ChevronLeft className="size-3.5" strokeWidth="2" />
+				</button>
+
+				{#each pages as item, index (item === 'ellipsis' ? `ellipsis-${index}` : item)}
+					{#if item === 'ellipsis'}
+						<span
+							class="inline-flex size-7 items-center justify-center text-xs text-gray-400 select-none dark:text-gray-600"
+						>
+							...
+						</span>
+					{:else}
+						<button
+							class="inline-flex size-7 items-center justify-center rounded-lg text-xs font-medium transition hover:bg-gray-50 hover:text-gray-700 dark:hover:bg-gray-850 dark:hover:text-gray-300 {item ===
+							page
+								? 'bg-gray-50 text-gray-800 dark:bg-gray-850 dark:text-gray-100'
+								: 'text-gray-500 dark:text-gray-500'}"
+							aria-label={`Page ${item}`}
+							aria-current={item === page ? 'page' : undefined}
+							on:click={() => onPageChange(item)}
+						>
+							{item}
+						</button>
+					{/if}
+				{/each}
+
+				<button
+					class="inline-flex size-7 items-center justify-center rounded-lg text-xs font-medium text-gray-500 transition hover:bg-gray-50 hover:text-gray-700 disabled:cursor-not-allowed disabled:opacity-25 dark:text-gray-500 dark:hover:bg-gray-850 dark:hover:text-gray-300"
+					disabled={chatListLoading || page >= totalPages}
+					aria-label={$i18n.t('Next page')}
+					on:click={() => onPageChange(page + 1)}
+				>
+					<ChevronRight className="size-3.5" strokeWidth="2" />
+				</button>
+			</div>
 		{/if}
 	</div>
 {/if}
diff --git a/src/lib/components/chat/Placeholder/FolderPlaceholder.svelte b/src/lib/components/chat/Placeholder/FolderPlaceholder.svelte
index d7cbc6b400..c943f57ee0 100644
--- a/src/lib/components/chat/Placeholder/FolderPlaceholder.svelte
+++ b/src/lib/components/chat/Placeholder/FolderPlaceholder.svelte
@@ -11,18 +11,21 @@
 	import ChatList from './ChatList.svelte';
 	import FolderKnowledge from './FolderKnowledge.svelte';
 	import Spinner from '$lib/components/common/Spinner.svelte';
-	import { getChatListByFolderId } from '$lib/apis/chats';
 	import { getSharedFolderChats } from '$lib/apis/folders';
 
 	export let folder: any = null;
 
 	let selectedTab = 'chats';
 
+	const CHATS_PAGE_SIZE = 10;
 	let page = 1;
+	let totalChats = 0;
+	let orderBy: 'title' | 'updated_at' = 'updated_at';
+	let direction: 'asc' | 'desc' = 'desc';
+	let currentFolderId: string | null = null;
 
 	let chats: any[] | null = null;
 	let chatListLoading = false;
-	let allChatsLoaded = false;
 
 	$: showOwnerInfo = Boolean(
 		folder?.shared ||
@@ -30,42 +33,68 @@
 		(folder?.access_grants?.length ?? 0) > 0
 	);
 
-	const loadChats = async () => {
-		// getSharedFolderChats returns all users' chats in one shot; no pagination
-		allChatsLoaded = true;
+	const setSortKey = (key: 'title' | 'updated_at') => {
+		if (orderBy === key) {
+			direction = direction === 'asc' ? 'desc' : 'asc';
+		} else {
+			orderBy = key;
+			direction = key === 'title' ? 'asc' : 'desc';
+		}
+		page = 1;
+		setChatList();
+	};
+
+	const setPage = (nextPage: number) => {
+		if (nextPage === page || chatListLoading) {
+			return;
+		}
+
+		page = nextPage;
+		setChatList();
 	};
 
 	const setChatList = async () => {
+		const folderId = folder?.id;
 		chats = null;
-		page = 1;
-		allChatsLoaded = false;
-		chatListLoading = false;
 
-		if (folder && folder.id) {
+		if (folderId) {
 			// Always use the shared folder endpoint so owners also see
 			// chats created by users who have write access to this folder.
-			const res = await getSharedFolderChats(localStorage.token, folder.id).catch((error) => {
+			chatListLoading = true;
+			const res = await getSharedFolderChats(localStorage.token, folderId, {
+				page,
+				sortBy: orderBy,
+				sortDir: direction
+			}).catch((error) => {
 				console.error(error);
 				return null;
 			});
+			chatListLoading = false;
+
 			if (res && res.chats) {
 				chats = res.chats;
-				allChatsLoaded = true;
+				totalChats = res.total ?? res.chats.length;
 			} else {
-				// Fallback to regular API (e.g. if user has no shared access)
-				const fallback = await getChatListByFolderId(localStorage.token, folder.id, page).catch(
-					() => []
-				);
-				chats = fallback || [];
+				chats = [];
+				totalChats = 0;
 			}
 		} else {
 			chats = [];
+			totalChats = 0;
 		}
 	};
 
-	$: if (folder) {
+	$: if (folder?.id && folder.id !== currentFolderId) {
+		currentFolderId = folder.id;
+		page = 1;
 		setChatList();
 	}
+
+	$: if (!folder?.id && currentFolderId !== null) {
+		currentFolderId = null;
+		chats = [];
+		totalChats = 0;
+	}
 </script>
 
 <div>
@@ -105,9 +134,14 @@
 				<ChatList
 					{chats}
 					{chatListLoading}
-					{allChatsLoaded}
-					loadHandler={loadChats}
 					{showOwnerInfo}
+					{page}
+					total={totalChats}
+					perPage={CHATS_PAGE_SIZE}
+					{orderBy}
+					{direction}
+					onPageChange={setPage}
+					onSort={setSortKey}
 				/>
 			{:else}
 				<div class="py-10">
diff --git a/src/lib/components/layout/Sidebar/RecursiveFolder.svelte b/src/lib/components/layout/Sidebar/RecursiveFolder.svelte
index c13058200c..f971a93a9c 100644
--- a/src/lib/components/layout/Sidebar/RecursiveFolder.svelte
+++ b/src/lib/components/layout/Sidebar/RecursiveFolder.svelte
@@ -44,7 +44,6 @@
 	import DeleteConfirmDialog from '$lib/components/common/ConfirmDialog.svelte';
 	import FolderModal from './Folders/FolderModal.svelte';
 	import Emoji from '$lib/components/common/Emoji.svelte';
-	import Spinner from '$lib/components/common/Spinner.svelte';
 
 	export let folderRegistry = {};
 	export let open = false;
@@ -385,24 +384,46 @@
 		}, 500);
 	};
 
+	const SIDEBAR_CHATS_PAGE_SIZE = 10;
+	/** @type {any[] | null} */
 	let chats = null;
-	export const setFolderItems = async () => {
+	let chatsPage = 1;
+	let hasMoreChats = false;
+	let chatsLoading = false;
+
+	export const setFolderItems = async (append = false) => {
 		await tick();
-		if (open) {
+		if (open && !chatsLoading) {
 			// Always use getSharedFolderChats so owners also see chats
 			// created by users who have write access to this folder.
+			const nextPage = append ? chatsPage + 1 : 1;
+			chatsLoading = true;
 			try {
-				const res = await getSharedFolderChats(localStorage.token, folderId);
-				chats = res?.chats ?? [];
+				const res = await getSharedFolderChats(localStorage.token, folderId, {
+					page: nextPage
+				});
+				const nextChats = res?.chats ?? [];
+				chats = append ? [...(chats ?? []), ...nextChats] : nextChats;
+				chatsPage = nextPage;
+				hasMoreChats = res?.has_more ?? nextChats.length === SIDEBAR_CHATS_PAGE_SIZE;
 			} catch (error) {
 				// Fallback to regular API
-				chats = await getChatListByFolderId(localStorage.token, folderId).catch((error) => {
-					toast.error(`${error}`);
-					return [];
-				});
+				const fallback = await getChatListByFolderId(localStorage.token, folderId, nextPage).catch(
+					(error) => {
+						toast.error(`${error}`);
+						return [];
+					}
+				);
+				chats = append ? [...(chats ?? []), ...(fallback ?? [])] : (fallback ?? []);
+				chatsPage = nextPage;
+				hasMoreChats = (fallback?.length ?? 0) === SIDEBAR_CHATS_PAGE_SIZE;
+			} finally {
+				chatsLoading = false;
 			}
 		} else {
 			chats = null;
+			chatsPage = 1;
+			hasMoreChats = false;
 		}
 	};
 
@@ -564,7 +585,7 @@
 							await selectedFolder.set({ ...folders[folderId], ...folder });
 						}
 
-						await goto('/');
+						await goto(`/folders/${folderId}`);
 
 						if ($mobile) {
 							showSidebar.set(!$showSidebar);
@@ -678,7 +699,7 @@
 		</div>
 
 		<div slot="content" class="w-full">
-			{#if (folders[folderId]?.childrenIds ?? []).length > 0 || (chats ?? []).length > 0}
+			{#if (folders[folderId]?.childrenIds ?? []).length > 0 || (chats ?? []).length > 0 || hasMoreChats}
 				<div
 					class="ml-3 pl-1 mt-[1px] flex flex-col overflow-y-auto scrollbar-hidden border-s border-gray-100 dark:border-gray-900"
 				>
@@ -731,12 +752,41 @@
 							}}
 						/>
 					{/each}
+
+					{#if hasMoreChats}
+						<button
+							class="w-full px-2 py-0.5 text-left text-[11px] text-gray-400 transition hover:text-gray-700 disabled:cursor-not-allowed dark:text-gray-600 dark:hover:text-gray-300"
+							disabled={chatsLoading}
+							on:click={() => setFolderItems(true)}
+						>
+							{#if chatsLoading}
+								<div class="flex gap-1 px-2 py-1.5" aria-label="Loading">
+									<span class="size-1 rounded-full bg-gray-400 animate-pulse dark:bg-gray-600"
+									></span>
+									<span
+										class="size-1 rounded-full bg-gray-400 animate-pulse [animation-delay:150ms] dark:bg-gray-600"
+									></span>
+									<span
+										class="size-1 rounded-full bg-gray-400 animate-pulse [animation-delay:300ms] dark:bg-gray-600"
+									></span>
+								</div>
+							{:else}
+								{$i18n.t('Show more')}
+							{/if}
+						</button>
+					{/if}
 				</div>
 			{/if}
 
-			{#if chats === null}
-				<div class="flex justify-center items-center p-2">
-					<Spinner className="size-4 text-gray-500" />
+			{#if chats === null && chatsLoading}
+				<div class="flex gap-1 px-2 py-1.5" aria-label="Loading">
+					<span class="size-1 rounded-full bg-gray-400 animate-pulse dark:bg-gray-600"></span>
+					<span
+						class="size-1 rounded-full bg-gray-400 animate-pulse [animation-delay:150ms] dark:bg-gray-600"
+					></span>
+					<span
+						class="size-1 rounded-full bg-gray-400 animate-pulse [animation-delay:300ms] dark:bg-gray-600"
+					></span>
 				</div>
 			{/if}
 		</div>
diff --git a/src/lib/components/layout/Sidebar/SharedFolderItem.svelte b/src/lib/components/layout/Sidebar/SharedFolderItem.svelte
index ba74ffbba4..98a92f9042 100644
--- a/src/lib/components/layout/Sidebar/SharedFolderItem.svelte
+++ b/src/lib/components/layout/Sidebar/SharedFolderItem.svelte
@@ -1,6 +1,8 @@
 <script lang="ts">
 	import { getContext, createEventDispatcher } from 'svelte';
-	const i18n = getContext('i18n');
+	import type { Writable } from 'svelte/store';
+
+	const i18n: Writable<any> = getContext('i18n');
 	const dispatch = createEventDispatcher();
 
 	import { goto } from '$app/navigation';
@@ -13,16 +15,18 @@
 	import ChevronRight from './icons/ChevronRight.svelte';
 	import Eye from './icons/Eye.svelte';
 	import FolderIcon from './icons/Folder.svelte';
-	import Spinner from '$lib/components/common/Spinner.svelte';
 
 	export let folder: any;
 	export let allSharedFolders: any[] = [];
 	export let className = '';
 
+	const SIDEBAR_CHATS_PAGE_SIZE = 10;
 	let expanded = false;
 	let chats: any[] = [];
 	let loading = false;
 	let loaded = false;
+	let page = 1;
+	let hasMoreChats = false;
 
 	$: children = allSharedFolders.filter((f) => f.parent_id === folder.id);
 	$: permission = folder.permission || 'read';
@@ -35,16 +39,27 @@
 		}
 	};
 
-	const loadChats = async () => {
+	const loadChats = async (append = false) => {
+		if (loading) {
+			return;
+		}
+
 		loading = true;
 		try {
-			const res = await getSharedFolderChats(localStorage.token, folder.id);
+			const nextPage = append ? page + 1 : 1;
+			const res = await getSharedFolderChats(localStorage.token, folder.id, {
+				page: nextPage
+			});
 			if (res) {
-				chats = res.chats || [];
+				const nextChats = res.chats || [];
+				chats = append ? [...chats, ...nextChats] : nextChats;
+				page = nextPage;
+				hasMoreChats = res.has_more ?? nextChats.length === SIDEBAR_CHATS_PAGE_SIZE;
 			}
 		} catch (e) {
 			console.error('Failed to load shared folder chats', e);
 			chats = [];
+			hasMoreChats = false;
 		} finally {
 			loading = false;
 			loaded = true;
@@ -92,10 +107,15 @@
 
 	{#if expanded}
 		<div class="pl-3">
-			{#if loading}
-				<div class="flex items-center gap-1.5 py-1 px-2 text-xs text-gray-400">
-					<Spinner className="size-3" />
-					<span>{$i18n.t('Loading...')}</span>
+			{#if loading && !loaded}
+				<div class="flex gap-1 px-2 py-1.5" aria-label="Loading">
+					<span class="size-1 rounded-full bg-gray-400 animate-pulse dark:bg-gray-600"></span>
+					<span
+						class="size-1 rounded-full bg-gray-400 animate-pulse [animation-delay:150ms] dark:bg-gray-600"
+					></span>
+					<span
+						class="size-1 rounded-full bg-gray-400 animate-pulse [animation-delay:300ms] dark:bg-gray-600"
+					></span>
 				</div>
 			{:else}
 				{#each chats as chat (chat.id)}
@@ -115,7 +135,29 @@
 					<svelte:self folder={child} {allSharedFolders} />
 				{/each}
 
-				{#if chats.length === 0 && children.length === 0}
+				{#if hasMoreChats}
+					<button
+						class="w-full px-2 py-0.5 text-left text-[11px] text-gray-400 transition hover:text-gray-700 disabled:cursor-not-allowed dark:text-gray-600 dark:hover:text-gray-300"
+						disabled={loading}
+						on:click={() => loadChats(true)}
+					>
+						{#if loading}
+							<div class="flex gap-1 px-2 py-1.5" aria-label="Loading">
+								<span class="size-1 rounded-full bg-gray-400 animate-pulse dark:bg-gray-600"></span>
+								<span
+									class="size-1 rounded-full bg-gray-400 animate-pulse [animation-delay:150ms] dark:bg-gray-600"
+								></span>
+								<span
+									class="size-1 rounded-full bg-gray-400 animate-pulse [animation-delay:300ms] dark:bg-gray-600"
+								></span>
+							</div>
+						{:else}
+							{$i18n.t('Show more')}
+						{/if}
+					</button>
+				{/if}
+
+				{#if chats.length === 0 && children.length === 0 && !hasMoreChats}
 					<div class="text-[11px] text-gray-400 dark:text-gray-600 py-1 px-2">
 						{$i18n.t('Empty')}
 					</div>
diff --git a/src/routes/(app)/folders/[folderId]/+page.svelte b/src/routes/(app)/folders/[folderId]/+page.svelte
new file mode 100644
index 0000000000..16223b28b2
--- /dev/null
+++ b/src/routes/(app)/folders/[folderId]/+page.svelte
@@ -0,0 +1,46 @@
+<script lang="ts">
+	import { goto } from '$app/navigation';
+	import { page } from '$app/stores';
+	import { onDestroy, onMount } from 'svelte';
+	import { toast } from 'svelte-sonner';
+
+	import Chat from '$lib/components/chat/Chat.svelte';
+	import Spinner from '$lib/components/common/Spinner.svelte';
+	import { getFolderById } from '$lib/apis/folders';
+	import { selectedFolder } from '$lib/stores';
+
+	let ready = false;
+
+	onMount(async () => {
+		const folderId = $page.params.folderId;
+		if (!folderId) {
+			await goto('/');
+			return;
+		}
+
+		const folder = await getFolderById(localStorage.token, folderId).catch((error) => {
+			toast.error(`${error}`);
+			return null;
+		});
+
+		if (!folder) {
+			await goto('/');
+			return;
+		}
+
+		await selectedFolder.set(folder);
+		ready = true;
+	});
+
+	onDestroy(() => {
+		selectedFolder.set(null);
+	});
+</script>
+
+{#if ready}
+	<Chat />
+{:else}
+	<div class="w-full h-screen flex items-center justify-center">
+		<Spinner />
+	</div>
+{/if}

From 9ca8cf528af1c49da3f0a2bc3c6ca95c1dedbcf5 Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Tue, 21 Jul 2026 14:16:51 -0400
Subject: [PATCH 02/29] refac

---
 .../components/admin/Settings/Models.svelte   |  48 +--
 .../Settings/Models/ModelDefaultsPanel.svelte | 242 ++++++++++++++
 .../Settings/Models/ModelSettingsModal.svelte | 295 ------------------
 3 files changed, 271 insertions(+), 314 deletions(-)
 create mode 100644 src/lib/components/admin/Settings/Models/ModelDefaultsPanel.svelte
 delete mode 100644 src/lib/components/admin/Settings/Models/ModelSettingsModal.svelte

diff --git a/src/lib/components/admin/Settings/Models.svelte b/src/lib/components/admin/Settings/Models.svelte
index 7bd287e1ea..b3bc8b677a 100644
--- a/src/lib/components/admin/Settings/Models.svelte
+++ b/src/lib/components/admin/Settings/Models.svelte
@@ -33,8 +33,8 @@
 	import ModelEditor from '$lib/components/workspace/Models/ModelEditor.svelte';
 	import { toast } from 'svelte-sonner';
 	import ConfirmDialog from '$lib/components/common/ConfirmDialog.svelte';
-	import ModelSettingsModal from './Models/ModelSettingsModal.svelte';
 	import ManageModelsModal from './Models/ManageModelsModal.svelte';
+	import ModelDefaultsPanel from './Models/ModelDefaultsPanel.svelte';
 	import ModelMenu from '$lib/components/admin/Settings/Models/ModelMenu.svelte';
 	import EllipsisHorizontal from '$lib/components/icons/EllipsisHorizontal.svelte';
 	import EyeSlash from '$lib/components/icons/EyeSlash.svelte';
@@ -47,7 +47,6 @@
 	import Download from '$lib/components/icons/Download.svelte';
 	import EllipsisVertical from '$lib/components/icons/EllipsisVertical.svelte';
 	import Wrench from '$lib/components/icons/Wrench.svelte';
-	import SettingsIcon from '$lib/components/icons/Settings.svelte';
 	import { WEBUI_API_BASE_URL } from '$lib/constants';
 	import { goto } from '$app/navigation';
 
@@ -83,11 +82,13 @@
 	let filteredModels = [];
 	let selectedModelId = null;
 
-	let showConfigModal = false;
 	let showManageModal = false;
 	let showResetModal = false;
 	let savingModelOrder = false;
+	let savingModelsSettings = false;
 	let modelOrderDirty = false;
+	let modelDefaultsPanel = null;
+	let modelDefaultsDirty = false;
 
 	let viewOption = ''; // '' = All, 'enabled', 'disabled', 'visible', 'hidden'
 	let tags: string[] = [];
@@ -322,6 +323,20 @@
 		savingModelOrder = false;
 	};
 
+	const saveModelsSettings = async () => {
+		savingModelsSettings = true;
+
+		if (modelOrderDirty) {
+			await saveModelOrder(modelOrderList);
+		}
+
+		if (modelDefaultsDirty) {
+			await modelDefaultsPanel?.save();
+		}
+
+		savingModelsSettings = false;
+	};
+
 	const saveModelDefaults = async (
 		nextDefaultModelIds: string[],
 		nextDefaultPinnedModelIds: string[],
@@ -649,7 +664,6 @@
 	}}
 />
 
-<ModelSettingsModal bind:show={showConfigModal} initHandler={init} />
 <ManageModelsModal bind:show={showManageModal} />
 
 {#if models !== null}
@@ -662,18 +676,6 @@
 						{filteredModels.length}
 					</span>
 				</h2>
-
-				<Tooltip content={$i18n.t('Settings')}>
-					<button
-						type="button"
-						class="flex h-6 w-6 items-center justify-center rounded-lg text-gray-400 transition-colors duration-75 hover:text-gray-700 dark:text-gray-600 dark:hover:text-gray-300"
-						on:click={() => {
-							showConfigModal = true;
-						}}
-					>
-						<SettingsIcon className="size-3.5" />
-					</button>
-				</Tooltip>
 			</div>
 
 			{#if $user?.role === 'admin'}
@@ -714,6 +716,12 @@
 			{/if}
 
 			<div class="flex min-h-0 flex-1 flex-col space-y-1">
+				<ModelDefaultsPanel
+					bind:this={modelDefaultsPanel}
+					bind:dirty={modelDefaultsDirty}
+					initHandler={init}
+				/>
+
 				<div class="flex h-8 shrink-0 items-center w-full gap-2">
 					<div class="flex min-w-0 flex-1 items-center">
 						<div class=" self-center ml-1 mr-3">
@@ -1105,13 +1113,15 @@
 					<button
 						class="flex items-center gap-2 px-3.5 py-1.5 text-sm font-normal bg-black hover:bg-gray-900 text-white dark:bg-white dark:text-black dark:hover:bg-gray-100 transition rounded-full disabled:cursor-not-allowed disabled:opacity-40"
 						type="button"
-						disabled={!modelOrderDirty || savingModelOrder}
+						disabled={(!modelOrderDirty && !modelDefaultsDirty) ||
+							savingModelOrder ||
+							savingModelsSettings}
 						on:click={async () => {
-							await saveModelOrder(modelOrderList);
+							await saveModelsSettings();
 						}}
 					>
 						{$i18n.t('Save')}
-						{#if savingModelOrder}
+						{#if savingModelOrder || savingModelsSettings}
 							<span class="shrink-0">
 								<Spinner />
 							</span>
diff --git a/src/lib/components/admin/Settings/Models/ModelDefaultsPanel.svelte b/src/lib/components/admin/Settings/Models/ModelDefaultsPanel.svelte
new file mode 100644
index 0000000000..11043ed411
--- /dev/null
+++ b/src/lib/components/admin/Settings/Models/ModelDefaultsPanel.svelte
@@ -0,0 +1,242 @@
+<script>
+	import { getContext, onMount, tick } from 'svelte';
+	import { toast } from 'svelte-sonner';
+
+	const i18n = getContext('i18n');
+
+	import { config as appConfig } from '$lib/stores';
+	import { DEFAULT_CAPABILITIES } from '$lib/constants';
+	import { getModelsConfig, setModelsConfig, setDefaultPromptSuggestions } from '$lib/apis/configs';
+	import { getBackendConfig } from '$lib/apis';
+
+	import AdvancedParams from '$lib/components/chat/Settings/Advanced/AdvancedParams.svelte';
+	import Capabilities from '$lib/components/workspace/Models/Capabilities.svelte';
+	import DefaultFeatures from '$lib/components/workspace/Models/DefaultFeatures.svelte';
+	import BuiltinTools from '$lib/components/workspace/Models/BuiltinTools.svelte';
+	import PromptSuggestions from '$lib/components/workspace/Models/PromptSuggestions.svelte';
+
+	export let initHandler = () => {};
+	export let dirty = false;
+
+	let config = null;
+	let modelIds = [];
+	let loading = false;
+	let expanded = false;
+	let showCapabilities = false;
+	let showParameters = false;
+	let showPromptSuggestions = false;
+	let savedSnapshot = '';
+
+	let defaultCapabilities = {};
+	let defaultFeatureIds = [];
+	let defaultParams = {};
+	let builtinTools = {};
+	let promptSuggestions = [];
+
+	$: configuredParams = Object.entries(defaultParams ?? {}).filter(
+		([_, value]) => value !== null && value !== '' && value !== undefined
+	);
+	$: enabledCapabilities = Object.entries(defaultCapabilities ?? {}).filter(([_, value]) => value);
+	$: availableFeatures = enabledCapabilities
+		.filter(([key]) => ['web_search', 'code_interpreter', 'image_generation'].includes(key))
+		.map(([key]) => key);
+
+	const getSnapshot = () =>
+		JSON.stringify({
+			defaultCapabilities,
+			defaultFeatureIds,
+			defaultParams: Object.fromEntries(configuredParams),
+			builtinTools,
+			promptSuggestions: promptSuggestions.filter((p) => p.content !== '')
+		});
+
+	const updateDirty = async () => {
+		await tick();
+		dirty = savedSnapshot !== '' && getSnapshot() !== savedSnapshot;
+	};
+
+	const init = async () => {
+		loading = true;
+		config = await getModelsConfig(localStorage.token);
+
+		modelIds = config?.MODEL_ORDER_LIST || [];
+
+		const savedMeta = config?.DEFAULT_MODEL_METADATA;
+		if (savedMeta && Object.keys(savedMeta).length > 0) {
+			defaultCapabilities = savedMeta.capabilities ?? { ...DEFAULT_CAPABILITIES };
+			defaultFeatureIds = savedMeta.defaultFeatureIds ?? [];
+			builtinTools = savedMeta.builtinTools ?? {};
+		} else {
+			defaultCapabilities = { ...DEFAULT_CAPABILITIES };
+			defaultFeatureIds = [];
+			builtinTools = {};
+		}
+
+		defaultParams = config?.DEFAULT_MODEL_PARAMS ?? {};
+		promptSuggestions = $appConfig?.default_prompt_suggestions ?? [];
+		savedSnapshot = getSnapshot();
+		dirty = false;
+		loading = false;
+	};
+
+	export const save = async () => {
+		if (loading || !dirty) {
+			return true;
+		}
+
+		const metadata = {
+			capabilities: defaultCapabilities,
+			...(defaultFeatureIds.length > 0 ? { defaultFeatureIds } : {}),
+			...(Object.keys(builtinTools).length > 0 ? { builtinTools } : {})
+		};
+
+		const res = await setModelsConfig(localStorage.token, {
+			DEFAULT_MODELS: config?.DEFAULT_MODELS ?? null,
+			DEFAULT_PINNED_MODELS: config?.DEFAULT_PINNED_MODELS ?? null,
+			MODEL_ORDER_LIST: modelIds,
+			DEFAULT_MODEL_METADATA: metadata,
+			DEFAULT_MODEL_PARAMS: Object.fromEntries(configuredParams)
+		}).catch((error) => {
+			toast.error(`${error}`);
+			return null;
+		});
+
+		if (res) {
+			config = res;
+			promptSuggestions = promptSuggestions.filter((p) => p.content !== '');
+			promptSuggestions = await setDefaultPromptSuggestions(localStorage.token, promptSuggestions);
+			await appConfig.set(await getBackendConfig());
+			savedSnapshot = getSnapshot();
+			dirty = false;
+
+			toast.success($i18n.t('Models configuration saved successfully'));
+			initHandler();
+			return true;
+		} else {
+			toast.error($i18n.t('Failed to save models configuration'));
+			return false;
+		}
+	};
+
+	onMount(async () => {
+		await init();
+	});
+</script>
+
+<div class="shrink-0">
+	<div class="flex items-center justify-between gap-4 py-0.5">
+		<button
+			class="min-w-0 flex-1 text-left text-xs text-gray-600 transition hover:text-gray-900 dark:text-gray-400 dark:hover:text-gray-100"
+			type="button"
+			on:click={() => {
+				expanded = !expanded;
+			}}
+		>
+			{$i18n.t('Model Defaults')}
+		</button>
+
+		<button
+			class="shrink-0 text-[0.6875rem] text-gray-400 transition hover:text-gray-700 dark:text-gray-600 dark:hover:text-gray-300"
+			type="button"
+			on:click={() => {
+				expanded = !expanded;
+			}}
+		>
+			{expanded ? $i18n.t('Close') : $i18n.t('Configure')}
+		</button>
+	</div>
+
+	{#if expanded}
+		{#if loading}
+			<div class="py-1 text-xs text-gray-400 dark:text-gray-600">{$i18n.t('Loading...')}</div>
+		{:else}
+			<div class="space-y-1 mt-0.5">
+				<div>
+					<button
+						class="flex w-full items-center justify-between gap-4 py-0.5 text-left"
+						type="button"
+						on:click={() => {
+							showCapabilities = !showCapabilities;
+						}}
+					>
+						<span class="text-xs text-gray-600 dark:text-gray-400">
+							{$i18n.t('Model Capabilities')}
+						</span>
+						<span class="text-[0.6875rem] text-gray-400 dark:text-gray-600">
+							{showCapabilities ? $i18n.t('Close') : $i18n.t('Configure')}
+						</span>
+					</button>
+
+					{#if showCapabilities}
+						<div class="pb-2" on:click={updateDirty} on:change={updateDirty}>
+							<Capabilities bind:capabilities={defaultCapabilities} />
+
+							{#if availableFeatures.length > 0}
+								<div class="mt-4">
+									<DefaultFeatures {availableFeatures} bind:featureIds={defaultFeatureIds} />
+								</div>
+							{/if}
+
+							{#if defaultCapabilities.builtin_tools}
+								<div class="mt-4">
+									<BuiltinTools bind:builtinTools />
+								</div>
+							{/if}
+						</div>
+					{/if}
+				</div>
+
+				<div>
+					<button
+						class="flex w-full items-center justify-between gap-4 py-0.5 text-left"
+						type="button"
+						on:click={() => {
+							showParameters = !showParameters;
+						}}
+					>
+						<span class="text-xs text-gray-600 dark:text-gray-400">
+							{$i18n.t('Model Parameters')}
+						</span>
+						<span class="text-[0.6875rem] text-gray-400 dark:text-gray-600">
+							{showParameters ? $i18n.t('Close') : $i18n.t('Configure')}
+						</span>
+					</button>
+
+					{#if showParameters}
+						<div
+							class="max-h-[24rem] overflow-y-auto pb-2 pr-1 scrollbar-hover"
+							on:click={updateDirty}
+							on:change={updateDirty}
+							on:input={updateDirty}
+						>
+							<AdvancedParams admin={true} custom={true} bind:params={defaultParams} />
+						</div>
+					{/if}
+				</div>
+
+				<div>
+					<button
+						class="flex w-full items-center justify-between gap-4 py-0.5 text-left"
+						type="button"
+						on:click={() => {
+							showPromptSuggestions = !showPromptSuggestions;
+						}}
+					>
+						<span class="text-xs text-gray-600 dark:text-gray-400">
+							{$i18n.t('Prompt Suggestions')}
+						</span>
+						<span class="text-[0.6875rem] text-gray-400 dark:text-gray-600">
+							{showPromptSuggestions ? $i18n.t('Close') : $i18n.t('Configure')}
+						</span>
+					</button>
+
+					{#if showPromptSuggestions}
+						<div class="pb-2" on:click={updateDirty} on:change={updateDirty} on:input={updateDirty}>
+							<PromptSuggestions bind:promptSuggestions />
+						</div>
+					{/if}
+				</div>
+			</div>
+		{/if}
+	{/if}
+</div>
diff --git a/src/lib/components/admin/Settings/Models/ModelSettingsModal.svelte b/src/lib/components/admin/Settings/Models/ModelSettingsModal.svelte
deleted file mode 100644
index df1caaeea0..0000000000
--- a/src/lib/components/admin/Settings/Models/ModelSettingsModal.svelte
+++ /dev/null
@@ -1,295 +0,0 @@
-<script>
-	import { toast } from 'svelte-sonner';
-
-	import { getContext, onMount } from 'svelte';
-	const i18n = getContext('i18n');
-
-	import { config as _config } from '$lib/stores';
-	import { DEFAULT_CAPABILITIES } from '$lib/constants';
-	import { getModelsConfig, setModelsConfig, setDefaultPromptSuggestions } from '$lib/apis/configs';
-	import { getBackendConfig } from '$lib/apis';
-
-	import Modal from '$lib/components/common/Modal.svelte';
-	import Spinner from '$lib/components/common/Spinner.svelte';
-	import ChevronUp from '$lib/components/icons/ChevronUp.svelte';
-	import ChevronDown from '$lib/components/icons/ChevronDown.svelte';
-	import XMark from '$lib/components/icons/XMark.svelte';
-	import AdvancedParams from '$lib/components/chat/Settings/Advanced/AdvancedParams.svelte';
-
-	import Capabilities from '$lib/components/workspace/Models/Capabilities.svelte';
-	import DefaultFeatures from '$lib/components/workspace/Models/DefaultFeatures.svelte';
-	import BuiltinTools from '$lib/components/workspace/Models/BuiltinTools.svelte';
-	import PromptSuggestions from '$lib/components/workspace/Models/PromptSuggestions.svelte';
-
-	import AdjustmentsHorizontal from '$lib/components/icons/AdjustmentsHorizontal.svelte';
-
-	export let show = false;
-	export let initHandler = () => {};
-
-	let config = null;
-
-	let modelIds = [];
-
-	let loading = false;
-	let showDefaultCapabilities = false;
-	let showDefaultParams = false;
-	let showDefaultPromptSuggestions = false;
-
-	let defaultCapabilities = {};
-	let defaultFeatureIds = [];
-	let defaultParams = {};
-	let builtinTools = {};
-	let promptSuggestions = [];
-
-	$: if (show) {
-		init();
-	}
-	const init = async () => {
-		config = await getModelsConfig(localStorage.token);
-
-		modelIds = config.MODEL_ORDER_LIST || [];
-
-		const savedMeta = config?.DEFAULT_MODEL_METADATA;
-		if (savedMeta && Object.keys(savedMeta).length > 0) {
-			defaultCapabilities = savedMeta.capabilities ?? { ...DEFAULT_CAPABILITIES };
-			defaultFeatureIds = savedMeta.defaultFeatureIds ?? [];
-			builtinTools = savedMeta.builtinTools ?? {};
-		} else {
-			defaultCapabilities = { ...DEFAULT_CAPABILITIES };
-			defaultFeatureIds = [];
-			builtinTools = {};
-		}
-		defaultParams = config?.DEFAULT_MODEL_PARAMS ?? {};
-
-		promptSuggestions = $_config?.default_prompt_suggestions ?? [];
-	};
-	const submitHandler = async () => {
-		loading = true;
-
-		const metadata = {
-			capabilities: defaultCapabilities,
-			...(defaultFeatureIds.length > 0 ? { defaultFeatureIds } : {}),
-			...(Object.keys(builtinTools).length > 0 ? { builtinTools } : {})
-		};
-
-		const res = await setModelsConfig(localStorage.token, {
-			DEFAULT_MODELS: config?.DEFAULT_MODELS ?? null,
-			DEFAULT_PINNED_MODELS: config?.DEFAULT_PINNED_MODELS ?? null,
-			MODEL_ORDER_LIST: modelIds,
-			DEFAULT_MODEL_METADATA: metadata,
-			DEFAULT_MODEL_PARAMS: Object.fromEntries(
-				Object.entries(defaultParams).filter(([_, v]) => v !== null && v !== '' && v !== undefined)
-			)
-		});
-
-		if (res) {
-			promptSuggestions = promptSuggestions.filter((p) => p.content !== '');
-			promptSuggestions = await setDefaultPromptSuggestions(localStorage.token, promptSuggestions);
-			await _config.set(await getBackendConfig());
-
-			toast.success($i18n.t('Models configuration saved successfully'));
-			initHandler();
-			show = false;
-		} else {
-			toast.error($i18n.t('Failed to save models configuration'));
-		}
-
-		loading = false;
-	};
-
-	onMount(async () => {
-		init();
-	});
-</script>
-
-<Modal size="lg" bind:show className="bg-white dark:bg-gray-900 rounded-4xl">
-	<div>
-		<div class=" flex justify-between dark:text-gray-100 px-4 pt-3 pb-1">
-			<div class=" text-sm font-medium self-center">
-				{$i18n.t('Settings')}
-			</div>
-			<button
-				class="self-center rounded-lg p-1 text-gray-500 transition hover:bg-gray-50 hover:text-gray-700 dark:text-gray-400 dark:hover:bg-gray-800 dark:hover:text-gray-200"
-				on:click={() => {
-					show = false;
-				}}
-			>
-				<XMark className={'size-4'} />
-			</button>
-		</div>
-
-		<div class="flex flex-col md:flex-row w-full px-4 pb-4 md:space-x-4 dark:text-gray-200">
-			<div class=" flex flex-col w-full sm:flex-row sm:justify-center sm:space-x-6">
-				{#if config}
-					<form
-						class="flex flex-col w-full"
-						on:submit|preventDefault={() => {
-							submitHandler();
-						}}
-					>
-						<div class="flex flex-col lg:flex-row w-full h-full pb-2 lg:space-x-4">
-							<div
-								id="admin-settings-tabs-container"
-								class="tabs flex flex-row overflow-x-auto gap-2.5 max-w-full lg:gap-1 lg:flex-col lg:flex-none lg:w-40 dark:text-gray-200 text-sm font-normal text-left scrollbar-none"
-							>
-								<button
-									class="px-0.5 py-1 max-w-fit w-fit rounded-lg flex-1 lg:flex-none flex text-right transition"
-									type="button"
-								>
-									<div class=" self-center mr-2">
-										<AdjustmentsHorizontal />
-									</div>
-									<div class=" self-center">{$i18n.t('Defaults')}</div>
-								</button>
-							</div>
-
-							<div class="flex-1 mt-1 lg:mt-1 lg:h-[30rem] lg:max-h-[30rem] flex flex-col min-w-0">
-								<div class="w-full h-full overflow-y-auto overflow-x-hidden scrollbar-hidden">
-									<div>
-										<button
-											class="flex w-full justify-between items-center"
-											type="button"
-											on:click={() => {
-												showDefaultPromptSuggestions = !showDefaultPromptSuggestions;
-											}}
-										>
-											<div class="text-xs text-gray-500 font-normal">
-												{$i18n.t('Prompt Suggestions')}
-											</div>
-											<div>
-												{#if showDefaultPromptSuggestions}
-													<ChevronUp className="size-3" />
-												{:else}
-													<ChevronDown className="size-3" />
-												{/if}
-											</div>
-										</button>
-
-										{#if showDefaultPromptSuggestions}
-											<div class="mt-2">
-												<PromptSuggestions bind:promptSuggestions />
-
-												{#if promptSuggestions.length > 0}
-													<div class="text-xs text-left w-full mt-2 text-gray-500">
-														{$i18n.t(
-															'Adjusting these settings will apply changes universally to all users.'
-														)}
-													</div>
-												{/if}
-											</div>
-										{/if}
-									</div>
-
-									<hr class=" border-gray-50 dark:border-gray-800/10 my-2.5 w-full" />
-
-									<div>
-										<button
-											class="flex w-full justify-between items-center"
-											type="button"
-											on:click={() => {
-												showDefaultCapabilities = !showDefaultCapabilities;
-											}}
-										>
-											<div class="text-xs text-gray-500 font-normal">
-												{$i18n.t('Model Capabilities')}
-											</div>
-											<div>
-												{#if showDefaultCapabilities}
-													<ChevronUp className="size-3" />
-												{:else}
-													<ChevronDown className="size-3" />
-												{/if}
-											</div>
-										</button>
-
-										{#if showDefaultCapabilities}
-											<div class="mt-2">
-												<Capabilities bind:capabilities={defaultCapabilities} />
-
-												{#if Object.keys(defaultCapabilities).filter((key) => defaultCapabilities[key]).length > 0}
-													{@const availableFeatures = Object.entries(defaultCapabilities)
-														.filter(
-															([key, value]) =>
-																value &&
-																['web_search', 'code_interpreter', 'image_generation'].includes(key)
-														)
-														.map(([key, value]) => key)}
-
-													{#if availableFeatures.length > 0}
-														<div class="mt-4">
-															<DefaultFeatures
-																{availableFeatures}
-																bind:featureIds={defaultFeatureIds}
-															/>
-														</div>
-													{/if}
-												{/if}
-
-												{#if defaultCapabilities.builtin_tools}
-													<div class="mt-4">
-														<BuiltinTools bind:builtinTools />
-													</div>
-												{/if}
-											</div>
-										{/if}
-									</div>
-
-									<hr class=" border-gray-50 dark:border-gray-800/10 my-2.5 w-full" />
-
-									<div>
-										<button
-											class="flex w-full justify-between items-center"
-											type="button"
-											on:click={() => {
-												showDefaultParams = !showDefaultParams;
-											}}
-										>
-											<div class="text-xs text-gray-500 font-normal">
-												{$i18n.t('Model Parameters')}
-											</div>
-											<div>
-												{#if showDefaultParams}
-													<ChevronUp className="size-3" />
-												{:else}
-													<ChevronDown className="size-3" />
-												{/if}
-											</div>
-										</button>
-
-										{#if showDefaultParams}
-											<div class="mt-2">
-												<AdvancedParams admin={true} custom={true} bind:params={defaultParams} />
-											</div>
-										{/if}
-									</div>
-								</div>
-
-								<div class="flex justify-end items-center pt-3 text-sm font-normal gap-1.5">
-									<button
-										class="px-3.5 py-1.5 text-sm font-normal bg-black hover:bg-gray-900 text-white dark:bg-white dark:text-black dark:hover:bg-gray-100 transition rounded-full flex items-center gap-2 whitespace-nowrap {loading
-											? ' cursor-not-allowed'
-											: ''}"
-										type="submit"
-										disabled={loading}
-									>
-										{$i18n.t('Save')}
-
-										{#if loading}
-											<span class="shrink-0">
-												<Spinner />
-											</span>
-										{/if}
-									</button>
-								</div>
-							</div>
-						</div>
-					</form>
-				{:else}
-					<div>
-						<Spinner className="size-5" />
-					</div>
-				{/if}
-			</div>
-		</div>
-	</div>
-</Modal>

From c4efa81d08c425678c810c51b4d62716e1e57117 Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Tue, 21 Jul 2026 14:18:41 -0400
Subject: [PATCH 03/29] refac

---
 .../components/admin/Settings/Models.svelte   | 78 +++++++++++++++++++
 1 file changed, 78 insertions(+)

diff --git a/src/lib/components/admin/Settings/Models.svelte b/src/lib/components/admin/Settings/Models.svelte
index b3bc8b677a..6fc41df487 100644
--- a/src/lib/components/admin/Settings/Models.svelte
+++ b/src/lib/components/admin/Settings/Models.svelte
@@ -40,6 +40,7 @@
 	import EyeSlash from '$lib/components/icons/EyeSlash.svelte';
 	import Eye from '$lib/components/icons/Eye.svelte';
 	import ChevronDown from '$lib/components/icons/ChevronDown.svelte';
+	import Check from '$lib/components/icons/Check.svelte';
 	import CheckCircle from '$lib/components/icons/CheckCircle.svelte';
 	import GarbageBin from '$lib/components/icons/GarbageBin.svelte';
 	import Minus from '$lib/components/icons/Minus.svelte';
@@ -47,6 +48,10 @@
 	import Download from '$lib/components/icons/Download.svelte';
 	import EllipsisVertical from '$lib/components/icons/EllipsisVertical.svelte';
 	import Wrench from '$lib/components/icons/Wrench.svelte';
+	import Pin from '$lib/components/icons/Pin.svelte';
+	import PinSlash from '$lib/components/icons/PinSlash.svelte';
+	import GlobeAlt from '$lib/components/icons/GlobeAlt.svelte';
+	import LockClosed from '$lib/components/icons/LockClosed.svelte';
 	import { WEBUI_API_BASE_URL } from '$lib/constants';
 	import { goto } from '$app/navigation';
 
@@ -1013,6 +1018,79 @@
 												{/if}
 											</button>
 										</Tooltip>
+
+										<Tooltip
+											content={defaultModelIdSet.has(model.id)
+												? $i18n.t('Remove Selected Model')
+												: $i18n.t('Set as Selected Model')}
+										>
+											<button
+												class="self-center w-fit text-sm p-1.5 rounded-xl hover:bg-black/5 dark:hover:bg-white/5 {defaultModelIdSet.has(
+													model.id
+												)
+													? 'text-gray-900 dark:text-white'
+													: 'text-gray-500 dark:text-gray-400 dark:hover:text-white'}"
+												type="button"
+												aria-label={defaultModelIdSet.has(model.id)
+													? $i18n.t('Remove Selected Model')
+													: $i18n.t('Set as Selected Model')}
+												on:click={() => {
+													toggleDefaultModelHandler(model);
+												}}
+											>
+												<Check className="size-3.5" />
+											</button>
+										</Tooltip>
+
+										<Tooltip
+											content={defaultPinnedModelIdSet.has(model.id)
+												? $i18n.t('Remove Pinned Model')
+												: $i18n.t('Set as Pinned Model')}
+										>
+											<button
+												class="self-center w-fit text-sm p-1.5 rounded-xl hover:bg-black/5 dark:hover:bg-white/5 {defaultPinnedModelIdSet.has(
+													model.id
+												)
+													? 'text-gray-900 dark:text-white'
+													: 'text-gray-500 dark:text-gray-400 dark:hover:text-white'}"
+												type="button"
+												aria-label={defaultPinnedModelIdSet.has(model.id)
+													? $i18n.t('Remove Pinned Model')
+													: $i18n.t('Set as Pinned Model')}
+												on:click={() => {
+													toggleDefaultPinnedModelHandler(model);
+												}}
+											>
+												{#if defaultPinnedModelIdSet.has(model.id)}
+													<PinSlash className="size-3.5" />
+												{:else}
+													<Pin className="size-3.5" />
+												{/if}
+											</button>
+										</Tooltip>
+
+										<Tooltip
+											content={isPublicModel(model)
+												? $i18n.t('Make Private')
+												: $i18n.t('Make Public')}
+										>
+											<button
+												class="self-center w-fit text-sm p-1.5 rounded-xl text-gray-500 hover:bg-black/5 dark:text-gray-400 dark:hover:bg-white/5 dark:hover:text-white"
+												type="button"
+												aria-label={isPublicModel(model)
+													? $i18n.t('Make Private')
+													: $i18n.t('Make Public')}
+												on:click={() => {
+													toggleModelPrivacyHandler(model);
+												}}
+											>
+												{#if isPublicModel(model)}
+													<LockClosed className="size-3.5" />
+												{:else}
+													<GlobeAlt className="size-3.5" />
+												{/if}
+											</button>
+										</Tooltip>
 									{/if}
 
 									<button

From 5418ac921bb8651f6a77a06f94cc8bebfaebe3db Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 02:33:09 -0400
Subject: [PATCH 04/29] refac

---
 README.md | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/README.md b/README.md
index bc504ce641..1342c8d838 100644
--- a/README.md
+++ b/README.md
@@ -93,11 +93,9 @@ Want to learn more about Open WebUI's features? Check out our [Open WebUI docume
 
 Open WebUI is the core, surrounded by companion apps and infrastructure that extend what your AI can do, where it can reach, and how you run it:
 
-- ⚡ **Open Terminal** ([open-webui/open-terminal](https://github.com/open-webui/open-terminal)): A self-hosted computing environment that plugs into Open WebUI, giving the AI a place to write code, run it, read output, fix errors, and iterate inside the chat.
+- 💻 **Open WebUI Computer** ([open-webui/computer](https://github.com/open-webui/computer)): A standalone, mobile-first computer and coding agent that runs on the machine you own. Files, terminal, and git in a browser tab, reachable from your phone. Connect it into Open WebUI as a model, or reach it from Telegram, WhatsApp, and more.
 
-- 🔒 **Terminals** · Enterprise ([open-webui/terminals](https://github.com/open-webui/terminals)): Per-user isolated containers with separate credentials, resource limits, and network rules. Automatic lifecycle management on Docker or Kubernetes.
-
-- 💻 **cptr** ([open-webui/computer](https://github.com/open-webui/computer)): A standalone, mobile-first computer and coding agent that runs on the machine you own. Files, terminal, and git in a browser tab, reachable from your phone. Connect it into Open WebUI as a model, or reach it from Telegram, WhatsApp, and more.
+- ⚡ **Open Terminal** and **Terminals (Enterprise)** ([open-webui/open-terminal](https://github.com/open-webui/open-terminal) & [open-webui/terminals](https://github.com/open-webui/terminals)): A self-hosted computing environment that plugs into Open WebUI, giving the AI a place to write code, run it, read output, fix errors, and iterate inside the chat. Terminals gives you per-user isolated containers with separate credentials, resource limits, and network rules. Automatic lifecycle management on Docker or Kubernetes.
 
 - 🔄 **oikb** ([open-webui/oikb](https://github.com/open-webui/oikb)): Feed your Knowledge Bases from 45+ sources (GitHub, Confluence, ServiceNow, Salesforce, Jira, Slack, SharePoint, Notion, and more), keeping the tools your team already uses continuously in sync.
 

From cf887b68ea58bcd1d8b035842c4ea35a5113ed8b Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 02:54:56 -0400
Subject: [PATCH 05/29] refac

---
 ...2c3d4e5f_add_current_message_id_to_chat.py | 139 ++++++++++++++++++
 backend/open_webui/models/chats.py            |  43 +++++-
 backend/open_webui/routers/chats.py           |  94 +++++++++++-
 backend/open_webui/utils/chat_fork.py         |  41 ++++++
 .../open_webui/utils/context_compaction.py    |  18 ++-
 5 files changed, 326 insertions(+), 9 deletions(-)
 create mode 100644 backend/open_webui/migrations/versions/9a1b2c3d4e5f_add_current_message_id_to_chat.py
 create mode 100644 backend/open_webui/utils/chat_fork.py

diff --git a/backend/open_webui/migrations/versions/9a1b2c3d4e5f_add_current_message_id_to_chat.py b/backend/open_webui/migrations/versions/9a1b2c3d4e5f_add_current_message_id_to_chat.py
new file mode 100644
index 0000000000..771cb55f52
--- /dev/null
+++ b/backend/open_webui/migrations/versions/9a1b2c3d4e5f_add_current_message_id_to_chat.py
@@ -0,0 +1,139 @@
+"""add current_message_id to chat
+
+Revision ID: 9a1b2c3d4e5f
+Revises: 856c5b02fb54
+Create Date: 2026-07-23 00:00:00.000000
+
+"""
+
+import json
+from typing import Sequence, Union
+
+import sqlalchemy as sa
+from alembic import op
+
+revision: str = '9a1b2c3d4e5f'
+down_revision: Union[str, None] = '856c5b02fb54'
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    conn = op.get_bind()
+    columns = [col['name'] for col in sa.inspect(conn).get_columns('chat')]
+    if 'current_message_id' not in columns:
+        op.add_column('chat', sa.Column('current_message_id', sa.Text(), nullable=True))
+
+    chat = sa.table(
+        'chat',
+        sa.column('id', sa.String()),
+        sa.column('chat', sa.Text()),
+        sa.column('current_message_id', sa.Text()),
+    )
+    chat_message = sa.table(
+        'chat_message',
+        sa.column('id', sa.Text()),
+        sa.column('chat_id', sa.Text()),
+        sa.column('parent_id', sa.Text()),
+        sa.column('created_at', sa.BigInteger()),
+    )
+
+    messages_by_chat: dict[str, dict[str, dict]] = {}
+    for row in conn.execute(
+        sa.select(
+            chat_message.c.chat_id,
+            chat_message.c.id,
+            chat_message.c.parent_id,
+            chat_message.c.created_at,
+        )
+    ):
+        values = row._mapping
+        chat_id = values['chat_id']
+        prefix = f'{chat_id}-'
+        message_id = values['id']
+        if message_id and message_id.startswith(prefix):
+            message_id = message_id[len(prefix) :]
+        if not message_id:
+            continue
+        parent_id = values['parent_id']
+        if parent_id and parent_id.startswith(prefix):
+            parent_id = parent_id[len(prefix) :]
+        messages_by_chat.setdefault(chat_id, {})[message_id] = {
+            'parent_id': parent_id,
+            'created_at': values['created_at'] or 0,
+        }
+
+    for row in conn.execute(sa.select(chat.c.id, chat.c.chat, chat.c.current_message_id)):
+        values = row._mapping
+        chat_id = values['id']
+        prefix = f'{chat_id}-'
+        chat_messages = messages_by_chat.get(chat_id, {})
+
+        chat_data = {}
+        if isinstance(values['chat'], dict):
+            chat_data = values['chat']
+        elif isinstance(values['chat'], str):
+            try:
+                parsed = json.loads(values['chat'])
+                chat_data = parsed if isinstance(parsed, dict) else {}
+            except (TypeError, ValueError, json.JSONDecodeError):
+                chat_data = {}
+
+        history = chat_data.get('history') if isinstance(chat_data.get('history'), dict) else {}
+        messages = history.get('messages') if isinstance(history.get('messages'), dict) else {}
+        if not messages and isinstance(chat_data.get('messages'), list):
+            messages = {
+                message['id']: message
+                for message in chat_data['messages']
+                if isinstance(message, dict) and message.get('id')
+            }
+        json_messages = {
+            message_id: {
+                'parent_id': message.get('parentId') if isinstance(message, dict) else None,
+                'created_at': message.get('timestamp', 0) if isinstance(message, dict) else 0,
+            }
+            for message_id, message in messages.items()
+        }
+        available_messages = chat_messages or json_messages
+        candidates = [
+            values['current_message_id'],
+            history.get('currentId'),
+            chat_data.get('currentId'),
+            chat_data.get('branchPointMessageId'),
+        ]
+
+        current_message_id = next(
+            (
+                candidate[len(prefix) :] if candidate.startswith(prefix) else candidate
+                for candidate in candidates
+                if candidate
+                and (
+                    not available_messages
+                    or (candidate[len(prefix) :] if candidate.startswith(prefix) else candidate) in available_messages
+                )
+            ),
+            None,
+        )
+
+        if not current_message_id and available_messages:
+            parent_ids = {
+                message['parent_id']
+                for message in available_messages.values()
+                if message.get('parent_id') in available_messages
+            }
+            leaf_ids = [message_id for message_id in available_messages if message_id not in parent_ids]
+            current_message_id = max(
+                leaf_ids or list(available_messages),
+                key=lambda message_id: available_messages[message_id].get('created_at') or 0,
+            )
+
+        if current_message_id and current_message_id != values['current_message_id']:
+            conn.execute(
+                sa.update(chat)
+                .where(chat.c.id == chat_id)
+                .values(current_message_id=current_message_id)
+            )
+
+
+def downgrade() -> None:
+    op.drop_column('chat', 'current_message_id')
diff --git a/backend/open_webui/models/chats.py b/backend/open_webui/models/chats.py
index 1d959f0bc6..4f2213b84d 100644
--- a/backend/open_webui/models/chats.py
+++ b/backend/open_webui/models/chats.py
@@ -62,6 +62,7 @@ class Chat(Base):  # database table mapping for chat entity
 
     tasks = Column(JSON, nullable=True)
     summary = Column(Text, nullable=True)
+    current_message_id = Column(Text, nullable=True)
 
     last_read_at = Column(BigInteger, nullable=True)
 
@@ -98,6 +99,7 @@ class ChatModel(BaseModel):
 
     tasks: list | None = None
     summary: str | None = None
+    current_message_id: str | None = None
 
     last_read_at: int | None = None
 
@@ -145,6 +147,7 @@ class ChatForm(BaseModel):
 class ChatImportForm(ChatForm):
     meta: dict | None = {}
     pinned: bool | None = False
+    current_message_id: str | None = None
     created_at: int | None = None
     updated_at: int | None = None
 
@@ -177,6 +180,7 @@ class ChatResponse(BaseModel):
 
     tasks: list | None = None
     summary: str | None = None
+    current_message_id: str | None = None
     context_usage: dict | None = None
 
 
@@ -280,6 +284,21 @@ class ChatTable:
         """Recursively remove null bytes from strings in dict/list structures."""
         return sanitize_data_for_db(obj)
 
+    def get_current_message_id(self, chat: dict | None) -> str | None:
+        chat = chat or {}
+        history = chat.get('history') if isinstance(chat.get('history'), dict) else {}
+        current_id = history.get('currentId') or chat.get('currentId') or chat.get('branchPointMessageId')
+        if current_id:
+            return current_id
+
+        messages = chat.get('messages')
+        if isinstance(messages, list):
+            for message in reversed(messages):
+                if isinstance(message, dict) and message.get('id'):
+                    return message['id']
+
+        return None
+
     def _sanitize_chat_row(self, chat_item):
         """
         Clean a Chat SQLAlchemy model's title + chat JSON,
@@ -375,6 +394,7 @@ class ChatTable:
                     'chat': self._clean_null_bytes(form_data.chat),
                     'folder_id': form_data.folder_id,
                     'meta': internal_meta or {},
+                    'current_message_id': self.get_current_message_id(form_data.chat),
                     'created_at': int(time.time()),
                     'updated_at': int(time.time()),
                     'last_read_at': int(time.time()),
@@ -388,8 +408,14 @@ class ChatTable:
 
             # Dual-write initial messages to chat_message table
             try:
-                history = form_data.chat.get('history', {})
-                messages = history.get('messages', {})
+                history = form_data.chat.get('history') if isinstance(form_data.chat.get('history'), dict) else {}
+                messages = history.get('messages') if isinstance(history.get('messages'), dict) else {}
+                if not messages and isinstance(form_data.chat.get('messages'), list):
+                    messages = {
+                        message.get('id'): message
+                        for message in form_data.chat['messages']
+                        if isinstance(message, dict) and message.get('id')
+                    }
                 for message_id, message in messages.items():
                     if isinstance(message, dict) and message.get('role'):
                         await ChatMessages.upsert_message(
@@ -458,6 +484,7 @@ class ChatTable:
                 'meta': form_data.meta,
                 'pinned': form_data.pinned,
                 'folder_id': form_data.folder_id,
+                'current_message_id': form_data.current_message_id or self.get_current_message_id(form_data.chat),
                 'created_at': (form_data.created_at if form_data.created_at else int(time.time())),
                 'updated_at': (form_data.updated_at if form_data.updated_at else int(time.time())),
             }
@@ -497,8 +524,14 @@ class ChatTable:
 
             # Dual-write messages to chat_message table
             for form_data, chat_obj in zip(chat_import_forms, chats):
-                history = form_data.chat.get('history', {})
-                messages = history.get('messages', {})
+                history = form_data.chat.get('history') if isinstance(form_data.chat.get('history'), dict) else {}
+                messages = history.get('messages') if isinstance(history.get('messages'), dict) else {}
+                if not messages and isinstance(form_data.chat.get('messages'), list):
+                    messages = {
+                        message.get('id'): message
+                        for message in form_data.chat['messages']
+                        if isinstance(message, dict) and message.get('id')
+                    }
                 for message_id, message in messages.items():
                     if isinstance(message, dict) and message.get('role'):
                         try:
@@ -530,6 +563,8 @@ class ChatTable:
 
                 chat_item.chat = self._clean_null_bytes(chat)
                 chat_item.title = self._clean_null_bytes(chat['title']) if 'title' in chat else 'New Chat'
+                if any(key in chat for key in ('history', 'messages', 'currentId', 'branchPointMessageId')):
+                    chat_item.current_message_id = self.get_current_message_id(chat)
 
                 if touch:
                     chat_item.updated_at = int(time.time())
diff --git a/backend/open_webui/routers/chats.py b/backend/open_webui/routers/chats.py
index 5a565407a6..8747c65a4f 100644
--- a/backend/open_webui/routers/chats.py
+++ b/backend/open_webui/routers/chats.py
@@ -38,6 +38,7 @@ from open_webui.tasks import has_active_tasks, stop_item_tasks
 from open_webui.utils.access_control import filter_allowed_access_grants, has_permission
 from open_webui.utils.access_control.folders import has_folder_access
 from open_webui.utils.auth import get_admin_user, get_verified_user
+from open_webui.utils.chat_fork import build_fork_history
 from open_webui.utils.context_compaction import compact_chat_branch, get_chat_context_usage
 from open_webui.utils.misc import get_message_list
 from open_webui.utils.models import get_all_models
@@ -1167,7 +1168,8 @@ async def compact_chat_by_id(
 
     history = (chat.chat or {}).get('history') or {}
     messages_map = await Chats.get_messages_map_by_chat_id(id)
-    message_list = get_message_list(messages_map or history.get('messages') or {}, history.get('currentId'))
+    current_message_id = chat.current_message_id or history.get('currentId')
+    message_list = get_message_list(messages_map or history.get('messages') or {}, current_message_id)
     model_id = (form_data.model if form_data else None) or next(
         (message.get('model') for message in reversed(message_list) if message.get('model')),
         None,
@@ -1564,6 +1566,96 @@ class CloneForm(BaseModel):
     title: str | None = None
 
 
+class ForkForm(BaseModel):
+    message_id: str | None = None
+
+
+@router.post('/{id}/fork', response_model=ChatResponse | None)
+async def fork_chat_by_id(
+    request: Request,
+    id: str,
+    form_data: ForkForm | None = None,
+    user=Depends(get_verified_user),
+    db: AsyncSession = Depends(get_async_session),
+):
+    await require_chat_import_permission(request, user, db)
+
+    chat = await Chats.get_chat_by_id_and_user_id(id, user.id, db=db)
+    if not chat:
+        raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail=ERROR_MESSAGES.DEFAULT())
+
+    if await has_active_tasks(request.app.state.redis, id):
+        raise HTTPException(
+            status_code=status.HTTP_409_CONFLICT,
+            detail='Wait for the current response to finish before forking.',
+        )
+
+    history = (chat.chat or {}).get('history') or {}
+    messages_map = await Chats.get_messages_map_by_chat_id(id) or history.get('messages') or {}
+    if any(
+        message.get('role') == 'assistant' and message.get('done') is False
+        for message in messages_map.values()
+        if isinstance(message, dict)
+    ):
+        raise HTTPException(
+            status_code=status.HTTP_409_CONFLICT,
+            detail='Wait for the current response to finish before forking.',
+        )
+
+    source_message_id = (form_data.message_id if form_data else None) or chat.current_message_id or history.get('currentId')
+    if not source_message_id:
+        raise HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail='chat has no messages to fork')
+
+    try:
+        fork_history, fork_messages = build_fork_history(messages_map, source_message_id)
+    except ValueError as exc:
+        detail = str(exc)
+        raise HTTPException(
+            status_code=status.HTTP_404_NOT_FOUND if detail == 'message not found' else status.HTTP_400_BAD_REQUEST,
+            detail=detail,
+        ) from exc
+
+    updated_chat = {**(chat.chat or {})}
+    updated_chat.pop('currentId', None)
+    updated_chat.update(
+        {
+            'originalChatId': chat.id,
+            'branchPointMessageId': source_message_id,
+            'title': f'{chat.title} (fork)',
+            'history': fork_history,
+            'messages': fork_messages,
+        }
+    )
+    meta = {
+        **(chat.meta or {}),
+        'forked_from': chat.id,
+        'forked_from_message_id': source_message_id,
+    }
+
+    fork = await Chats.insert_new_chat(
+        str(uuid4()),
+        user.id,
+        ChatForm(chat=updated_chat, folder_id=chat.folder_id),
+        db=db,
+        internal_meta=meta,
+    )
+
+    if not fork:
+        raise HTTPException(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, detail=ERROR_MESSAGES.DEFAULT())
+
+    if chat.pinned:
+        fork = await Chats.toggle_chat_pinned_by_id(fork.id, db=db) or fork
+
+    await publish_event(
+        request,
+        EVENTS.CHAT_CLONED,
+        actor=user,
+        subject_id=fork.id,
+        data={'original_chat_id': id, 'forked_from_message_id': source_message_id},
+    )
+    return ChatResponse(**fork.model_dump())
+
+
 @router.post('/{id}/clone', response_model=ChatResponse | None)
 async def clone_chat_by_id(
     request: Request,
diff --git a/backend/open_webui/utils/chat_fork.py b/backend/open_webui/utils/chat_fork.py
new file mode 100644
index 0000000000..4963af0609
--- /dev/null
+++ b/backend/open_webui/utils/chat_fork.py
@@ -0,0 +1,41 @@
+from copy import deepcopy
+
+
+def build_fork_history(messages_map: dict, source_message_id: str) -> tuple[dict, list[dict]]:
+    if not messages_map:
+        raise ValueError('chat has no messages to fork')
+
+    branch: list[tuple[str, dict]] = []
+    seen: set[str] = set()
+    message_id = source_message_id
+
+    while message_id:
+        if message_id in seen:
+            raise ValueError('message branch contains a cycle')
+        seen.add(message_id)
+
+        message = messages_map.get(message_id)
+        if not isinstance(message, dict):
+            raise ValueError('message not found')
+
+        branch.append((message_id, message))
+        message_id = message.get('parentId')
+
+    fork_messages: dict[str, dict] = {}
+    ordered_messages: list[dict] = []
+    parent_id = None
+
+    for message_id, message in reversed(branch):
+        copied = deepcopy(message)
+        copied['id'] = message_id
+        copied['parentId'] = parent_id
+        copied['childrenIds'] = []
+
+        if parent_id:
+            fork_messages[parent_id]['childrenIds'] = [message_id]
+
+        fork_messages[message_id] = copied
+        ordered_messages.append(copied)
+        parent_id = message_id
+
+    return {'messages': fork_messages, 'currentId': source_message_id}, ordered_messages
diff --git a/backend/open_webui/utils/context_compaction.py b/backend/open_webui/utils/context_compaction.py
index ca0e7a4f9d..910fa5b7b6 100644
--- a/backend/open_webui/utils/context_compaction.py
+++ b/backend/open_webui/utils/context_compaction.py
@@ -148,8 +148,13 @@ async def compact_chat_branch(request, user, chat: Any, model_id: str, models: d
     if not config['enable']:
         return {'ok': True, 'compacted': False, 'reason': 'disabled'}
 
-    history = (chat.chat or {}).get('history') or {}
-    current_id = history.get('currentId')
+    chat_data = chat.chat or {}
+    history = chat_data.get('history') or {}
+    current_id = getattr(chat, 'current_message_id', None) or history.get('currentId')
+    if not current_id:
+        current_id = chat_data.get('currentId') or chat_data.get('branchPointMessageId')
+    if not current_id and isinstance(chat_data.get('messages'), list) and chat_data['messages']:
+        current_id = chat_data['messages'][-1].get('id')
     if not current_id:
         return {'ok': True, 'compacted': False, 'reason': 'empty'}
 
@@ -216,8 +221,13 @@ def _resolve_token_threshold(global_threshold: int, global_cap: int, metadata: d
 
 
 async def get_chat_context_usage(chat: Any, model_id: str | None = None) -> dict | None:
-    history = (chat.chat or {}).get('history') or {}
-    current_id = history.get('currentId')
+    chat_data = chat.chat or {}
+    history = chat_data.get('history') or {}
+    current_id = getattr(chat, 'current_message_id', None) or history.get('currentId')
+    if not current_id:
+        current_id = chat_data.get('currentId') or chat_data.get('branchPointMessageId')
+    if not current_id and isinstance(chat_data.get('messages'), list) and chat_data['messages']:
+        current_id = chat_data['messages'][-1].get('id')
     if not current_id:
         return None
 

From 73421c5b42ac5c2ebc5faa520b7ed3fa0e39f10d Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 02:55:06 -0400
Subject: [PATCH 06/29] refac

---
 src/lib/apis/chats/index.ts                   | 41 +++++++++++++++++++
 src/lib/components/chat/Chat.svelte           | 16 ++++----
 .../Commands/SlashCommands.svelte             |  6 +--
 src/lib/components/chat/Messages.svelte       |  2 +
 .../components/chat/Messages/Message.svelte   |  3 ++
 .../Messages/MultiResponseMessages.svelte     |  3 ++
 .../chat/Messages/ResponseMessage.svelte      | 32 +++++++++++++++
 src/lib/utils/index.ts                        | 21 ++++++----
 8 files changed, 103 insertions(+), 21 deletions(-)

diff --git a/src/lib/apis/chats/index.ts b/src/lib/apis/chats/index.ts
index b4bc2e002c..64d2960dbf 100644
--- a/src/lib/apis/chats/index.ts
+++ b/src/lib/apis/chats/index.ts
@@ -920,6 +920,47 @@ export const cloneChatById = async (token: string, id: string, title?: string) =
 	return res;
 };
 
+export const forkChatById = async (token: string, id: string, messageId?: string | null) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/chats/${id}/fork`, {
+		method: 'POST',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			...(token && { authorization: `Bearer ${token}` })
+		},
+		body: JSON.stringify({
+			message_id: messageId ?? null
+		})
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err;
+
+			if ('detail' in err) {
+				error = err.detail;
+			} else {
+				error = err;
+			}
+
+			console.error(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
 export const cloneSharedChatById = async (token: string, id: string) => {
 	let error = null;
 
diff --git a/src/lib/components/chat/Chat.svelte b/src/lib/components/chat/Chat.svelte
index 08bde0bc9f..d059846625 100644
--- a/src/lib/components/chat/Chat.svelte
+++ b/src/lib/components/chat/Chat.svelte
@@ -70,10 +70,10 @@
 
 	import {
 		archiveChatById,
-		cloneChatById,
 		compactChatById,
 		createNewChat,
 		deleteChatById,
+		forkChatById,
 		getAllTags,
 		getChatById,
 		getTagsById,
@@ -1884,6 +1884,9 @@
 					(chatContent?.history ?? undefined) !== undefined
 						? chatContent.history
 						: convertMessagesToHistory(chatContent.messages);
+				if (chat?.current_message_id && history?.messages?.[chat.current_message_id]) {
+					history.currentId = chat.current_message_id;
+				}
 
 				// Sanitize history: repair orphaned references and structurally-malformed
 				// nodes from failed regenerations (#24424, #24157, #20474)
@@ -2466,7 +2469,7 @@
 		document.getElementById('chat-input')?.focus();
 	};
 
-	const handleForkChat = async () => {
+	const handleForkChat = async (messageId: string | null = null) => {
 		if (!$chatId || !history?.currentId) {
 			toast.message($i18n.t('No chat to fork'));
 			return;
@@ -2489,13 +2492,7 @@
 		const toastId = toast.loading($i18n.t('Forking chat...'));
 
 		try {
-			const result = await cloneChatById(
-				localStorage.token,
-				$chatId,
-				$i18n.t('Clone of {{TITLE}}', {
-					TITLE: $chatTitle || 'Chat'
-				})
-			);
+			const result = await forkChatById(localStorage.token, $chatId, messageId ?? history.currentId);
 
 			if (result?.id) {
 				if (!embedded) {
@@ -3725,6 +3722,7 @@
 										{mergeResponses}
 										{chatActionHandler}
 										{addMessages}
+										forkHandler={generating || taskIds?.length ? null : handleForkChat}
 										topPadding={!embedded}
 										bottomPadding={files.length > 0}
 										{onSelect}
diff --git a/src/lib/components/chat/MessageInput/Commands/SlashCommands.svelte b/src/lib/components/chat/MessageInput/Commands/SlashCommands.svelte
index 084379ac90..75ec04ee1a 100644
--- a/src/lib/components/chat/MessageInput/Commands/SlashCommands.svelte
+++ b/src/lib/components/chat/MessageInput/Commands/SlashCommands.svelte
@@ -183,10 +183,10 @@
 				</button>
 			</Tooltip>
 		{:else if item.data.id === 'fork'}
-			<Tooltip content="Clone this chat into a new copy." placement="top">
+			<Tooltip content="Fork the current chat branch into a new chat." placement="top">
 				<button
 					type="button"
-					aria-label="Fork: clone this chat into a new copy."
+					aria-label="Fork: fork the current chat branch into a new chat."
 					class="slash-command-row flex items-center gap-2 w-full h-6 px-2 rounded-xl text-xs text-left transition-colors duration-75
 						{commandIdx === selectedIdx ? 'app-interactive-active' : ''} disabled:opacity-50"
 					disabled={forkDisabled}
@@ -222,7 +222,7 @@
 					</span>
 					<span class="flex-1 min-w-0 flex items-baseline gap-1.5 overflow-hidden">
 						<span class="truncate">Fork</span>
-						<span class="app-muted text-[0.625rem] truncate shrink-0">Clone chat</span>
+						<span class="app-muted text-[0.625rem] truncate shrink-0">Current branch</span>
 					</span>
 				</button>
 			</Tooltip>
diff --git a/src/lib/components/chat/Messages.svelte b/src/lib/components/chat/Messages.svelte
index 9322175b7c..0adb4a027a 100644
--- a/src/lib/components/chat/Messages.svelte
+++ b/src/lib/components/chat/Messages.svelte
@@ -40,6 +40,7 @@
 	export let showMessage: Function = () => {};
 	export let submitMessage: Function = () => {};
 	export let addMessages: Function = () => {};
+	export let forkHandler: Function | null = null;
 
 	export let readOnly = false;
 	export let editCodeBlock = true;
@@ -539,6 +540,7 @@
 								{continueResponse}
 								{mergeResponses}
 								{addMessages}
+								{forkHandler}
 								{triggerScroll}
 								{readOnly}
 								{editCodeBlock}
diff --git a/src/lib/components/chat/Messages/Message.svelte b/src/lib/components/chat/Messages/Message.svelte
index 5fb9a399da..0b18ca2d9f 100644
--- a/src/lib/components/chat/Messages/Message.svelte
+++ b/src/lib/components/chat/Messages/Message.svelte
@@ -39,6 +39,7 @@
 	export let mergeResponses;
 
 	export let addMessages;
+	export let forkHandler: Function | null = null;
 	export let triggerScroll;
 	export let readOnly = false;
 	export let editCodeBlock = true;
@@ -97,6 +98,7 @@
 				{continueResponse}
 				{regenerateResponse}
 				{addMessages}
+				{forkHandler}
 				{readOnly}
 				{editCodeBlock}
 				{topPadding}
@@ -122,6 +124,7 @@
 					{mergeResponses}
 					{triggerScroll}
 					{addMessages}
+					{forkHandler}
 					{readOnly}
 					{editCodeBlock}
 					{topPadding}
diff --git a/src/lib/components/chat/Messages/MultiResponseMessages.svelte b/src/lib/components/chat/Messages/MultiResponseMessages.svelte
index a58f6bd9a9..6223cf8338 100644
--- a/src/lib/components/chat/Messages/MultiResponseMessages.svelte
+++ b/src/lib/components/chat/Messages/MultiResponseMessages.svelte
@@ -45,6 +45,7 @@
 	export let mergeResponses: Function;
 
 	export let addMessages: Function;
+	export let forkHandler: Function | null = null;
 
 	export let triggerScroll: Function;
 
@@ -319,6 +320,7 @@
 											groupedMessageIds[selectedModelIdx].messageIds.length - 1;
 									}}
 									{addMessages}
+									{forkHandler}
 									{readOnly}
 									{topPadding}
 									{onInsertToNote}
@@ -376,6 +378,7 @@
 												groupedMessageIds[modelIdx].messageIds.length - 1;
 										}}
 										{addMessages}
+										{forkHandler}
 										{readOnly}
 										{editCodeBlock}
 										{topPadding}
diff --git a/src/lib/components/chat/Messages/ResponseMessage.svelte b/src/lib/components/chat/Messages/ResponseMessage.svelte
index 11b71eaae0..36d05bc37b 100644
--- a/src/lib/components/chat/Messages/ResponseMessage.svelte
+++ b/src/lib/components/chat/Messages/ResponseMessage.svelte
@@ -159,6 +159,7 @@
 	export let submitMessage: Function;
 	export let continueResponse: Function;
 	export let regenerateResponse: Function;
+	export let forkHandler: Function | null = null;
 
 	export let addMessages: Function;
 
@@ -1054,6 +1055,37 @@
 									</button>
 								</Tooltip>
 
+								{#if message.done && !readOnly && forkHandler}
+									<Tooltip content="Fork chat" placement="bottom">
+										<button
+											aria-label="Fork chat"
+											class="{isLastMessage || ($settings?.highContrastMode ?? false)
+												? 'visible'
+												: 'invisible group-hover:visible'} p-1.5 hover:bg-black/5 dark:hover:bg-white/5 rounded-lg dark:hover:text-white hover:text-black transition"
+											on:click={() => {
+												forkHandler?.(message.id);
+											}}
+										>
+											<svg
+												class="w-4 h-4"
+												viewBox="0 0 24 24"
+												fill="none"
+												stroke="currentColor"
+												stroke-width="1.8"
+												stroke-linecap="round"
+												stroke-linejoin="round"
+												aria-hidden="true"
+											>
+												<path d="M4 12H9" />
+												<path d="M9 12C12.5 12 12.5 7 16 7H20" />
+												<path d="M17 4L20 7L17 10" />
+												<path d="M9 12C12.5 12 12.5 17 16 17H20" />
+												<path d="M17 14L20 17L17 20" />
+											</svg>
+										</button>
+									</Tooltip>
+								{/if}
+
 								{#if onInsertToNote && visibleResponseContent}
 									<Tooltip content={$i18n.t('Insert into note')} placement="bottom">
 										<button
diff --git a/src/lib/utils/index.ts b/src/lib/utils/index.ts
index 43cb79ddf7..eb23447b03 100644
--- a/src/lib/utils/index.ts
+++ b/src/lib/utils/index.ts
@@ -203,25 +203,28 @@ export const convertMessagesToHistory = (messages) => {
 	let messageId = null;
 
 	for (const message of messages) {
-		messageId = uuidv4();
-
-		if (parentMessageId !== null) {
-			history.messages[parentMessageId].childrenIds = [
-				...history.messages[parentMessageId].childrenIds,
-				messageId
-			];
-		}
+		messageId = message?.id ?? uuidv4();
+		const parentId = message?.parentId ?? parentMessageId;
 
 		history.messages[messageId] = {
 			...message,
 			id: messageId,
-			parentId: parentMessageId,
+			parentId,
 			childrenIds: []
 		};
 
 		parentMessageId = messageId;
 	}
 
+	for (const message of Object.values(history.messages)) {
+		if (message.parentId && history.messages[message.parentId]) {
+			history.messages[message.parentId].childrenIds = [
+				...history.messages[message.parentId].childrenIds,
+				message.id
+			];
+		}
+	}
+
 	history.currentId = messageId;
 	return history;
 };

From 060d5da473107e7367f00d0c7e674818878fb0a1 Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 03:01:01 -0400
Subject: [PATCH 07/29] refac

---
 ...2c3d4e5f_add_current_message_id_to_chat.py | 254 ++++++++++++------
 1 file changed, 169 insertions(+), 85 deletions(-)

diff --git a/backend/open_webui/migrations/versions/9a1b2c3d4e5f_add_current_message_id_to_chat.py b/backend/open_webui/migrations/versions/9a1b2c3d4e5f_add_current_message_id_to_chat.py
index 771cb55f52..403a51a5ef 100644
--- a/backend/open_webui/migrations/versions/9a1b2c3d4e5f_add_current_message_id_to_chat.py
+++ b/backend/open_webui/migrations/versions/9a1b2c3d4e5f_add_current_message_id_to_chat.py
@@ -17,10 +17,13 @@ down_revision: Union[str, None] = '856c5b02fb54'
 branch_labels: Union[str, Sequence[str], None] = None
 depends_on: Union[str, Sequence[str], None] = None
 
+BATCH_SIZE = 150
+
 
 def upgrade() -> None:
     conn = op.get_bind()
-    columns = [col['name'] for col in sa.inspect(conn).get_columns('chat')]
+    inspector = sa.inspect(conn)
+    columns = [col['name'] for col in inspector.get_columns('chat')]
     if 'current_message_id' not in columns:
         op.add_column('chat', sa.Column('current_message_id', sa.Text(), nullable=True))
 
@@ -38,100 +41,181 @@ def upgrade() -> None:
         sa.column('created_at', sa.BigInteger()),
     )
 
-    messages_by_chat: dict[str, dict[str, dict]] = {}
-    for row in conn.execute(
-        sa.select(
-            chat_message.c.chat_id,
-            chat_message.c.id,
-            chat_message.c.parent_id,
-            chat_message.c.created_at,
+    has_chat_message = 'chat_message' in inspector.get_table_names()
+    result = conn.execute(
+        sa.select(chat.c.id, chat.c.chat, chat.c.current_message_id).execution_options(
+            yield_per=BATCH_SIZE,
+            stream_results=True,
         )
-    ):
-        values = row._mapping
-        chat_id = values['chat_id']
-        prefix = f'{chat_id}-'
-        message_id = values['id']
-        if message_id and message_id.startswith(prefix):
-            message_id = message_id[len(prefix) :]
-        if not message_id:
-            continue
-        parent_id = values['parent_id']
-        if parent_id and parent_id.startswith(prefix):
-            parent_id = parent_id[len(prefix) :]
-        messages_by_chat.setdefault(chat_id, {})[message_id] = {
-            'parent_id': parent_id,
-            'created_at': values['created_at'] or 0,
-        }
+    )
 
-    for row in conn.execute(sa.select(chat.c.id, chat.c.chat, chat.c.current_message_id)):
-        values = row._mapping
-        chat_id = values['id']
-        prefix = f'{chat_id}-'
-        chat_messages = messages_by_chat.get(chat_id, {})
+    while True:
+        rows = result.fetchmany(BATCH_SIZE)
+        if not rows:
+            break
 
-        chat_data = {}
-        if isinstance(values['chat'], dict):
-            chat_data = values['chat']
-        elif isinstance(values['chat'], str):
-            try:
-                parsed = json.loads(values['chat'])
-                chat_data = parsed if isinstance(parsed, dict) else {}
-            except (TypeError, ValueError, json.JSONDecodeError):
-                chat_data = {}
+        batch_chat_ids: list[str] = []
+        candidates_by_chat: dict[str, list[str]] = {}
+        current_by_chat: dict[str, str | None] = {}
+        json_messages_by_chat: dict[str, dict[str, dict]] = {}
 
-        history = chat_data.get('history') if isinstance(chat_data.get('history'), dict) else {}
-        messages = history.get('messages') if isinstance(history.get('messages'), dict) else {}
-        if not messages and isinstance(chat_data.get('messages'), list):
-            messages = {
-                message['id']: message
-                for message in chat_data['messages']
-                if isinstance(message, dict) and message.get('id')
-            }
-        json_messages = {
-            message_id: {
-                'parent_id': message.get('parentId') if isinstance(message, dict) else None,
-                'created_at': message.get('timestamp', 0) if isinstance(message, dict) else 0,
-            }
-            for message_id, message in messages.items()
-        }
-        available_messages = chat_messages or json_messages
-        candidates = [
-            values['current_message_id'],
-            history.get('currentId'),
-            chat_data.get('currentId'),
-            chat_data.get('branchPointMessageId'),
-        ]
+        for row in rows:
+            values = row._mapping
+            chat_id = values['id']
+            prefix = f'{chat_id}-'
+            batch_chat_ids.append(chat_id)
+            current_by_chat[chat_id] = values['current_message_id']
 
-        current_message_id = next(
-            (
-                candidate[len(prefix) :] if candidate.startswith(prefix) else candidate
+            chat_data = {}
+            if isinstance(values['chat'], dict):
+                chat_data = values['chat']
+            elif isinstance(values['chat'], str):
+                try:
+                    parsed = json.loads(values['chat'])
+                    chat_data = parsed if isinstance(parsed, dict) else {}
+                except (TypeError, ValueError, json.JSONDecodeError):
+                    pass
+
+            history = chat_data.get('history') if isinstance(chat_data.get('history'), dict) else {}
+            candidates_by_chat[chat_id] = []
+            for candidate in (
+                values['current_message_id'],
+                history.get('currentId'),
+                chat_data.get('currentId'),
+                chat_data.get('branchPointMessageId'),
+            ):
+                if not isinstance(candidate, str) or not candidate:
+                    continue
+                candidate = candidate[len(prefix) :] if candidate.startswith(prefix) else candidate
+                if candidate not in candidates_by_chat[chat_id]:
+                    candidates_by_chat[chat_id].append(candidate)
+
+            messages = history.get('messages') if isinstance(history.get('messages'), dict) else {}
+            if not messages and isinstance(chat_data.get('messages'), list):
+                messages = {
+                    message['id']: message
+                    for message in chat_data['messages']
+                    if isinstance(message, dict) and message.get('id')
+                }
+            if messages:
+                json_messages_by_chat[chat_id] = {
+                    message_id: {
+                        'parent_id': message.get('parentId') if isinstance(message, dict) else None,
+                        'created_at': message.get('timestamp', 0) if isinstance(message, dict) else 0,
+                    }
+                    for message_id, message in messages.items()
+                }
+
+        resolved: dict[str, str] = {}
+
+        if has_chat_message:
+            candidate_ids = {
+                f'{chat_id}-{candidate}'
+                for chat_id, candidates in candidates_by_chat.items()
                 for candidate in candidates
-                if candidate
-                and (
-                    not available_messages
-                    or (candidate[len(prefix) :] if candidate.startswith(prefix) else candidate) in available_messages
-                )
-            ),
-            None,
-        )
-
-        if not current_message_id and available_messages:
-            parent_ids = {
-                message['parent_id']
-                for message in available_messages.values()
-                if message.get('parent_id') in available_messages
             }
-            leaf_ids = [message_id for message_id in available_messages if message_id not in parent_ids]
-            current_message_id = max(
-                leaf_ids or list(available_messages),
-                key=lambda message_id: available_messages[message_id].get('created_at') or 0,
-            )
+            if candidate_ids:
+                valid_by_chat: dict[str, set[str]] = {}
+                for row in conn.execute(
+                    sa.select(chat_message.c.chat_id, chat_message.c.id).where(
+                        chat_message.c.chat_id.in_(batch_chat_ids),
+                        chat_message.c.id.in_(candidate_ids),
+                    )
+                ):
+                    values = row._mapping
+                    chat_id = values['chat_id']
+                    prefix = f'{chat_id}-'
+                    message_id = values['id']
+                    if message_id and message_id.startswith(prefix):
+                        message_id = message_id[len(prefix) :]
+                    if message_id:
+                        valid_by_chat.setdefault(chat_id, set()).add(message_id)
+                for chat_id, candidates in candidates_by_chat.items():
+                    valid_ids = valid_by_chat.get(chat_id, set())
+                    for candidate in candidates:
+                        if candidate in valid_ids:
+                            resolved[chat_id] = candidate
+                            break
 
-        if current_message_id and current_message_id != values['current_message_id']:
+            unresolved_chat_ids = [chat_id for chat_id in batch_chat_ids if chat_id not in resolved]
+            messages_by_chat: dict[str, dict[str, dict]] = {}
+            if unresolved_chat_ids:
+                for row in conn.execute(
+                    sa.select(
+                        chat_message.c.chat_id,
+                        chat_message.c.id,
+                        chat_message.c.parent_id,
+                        chat_message.c.created_at,
+                    ).where(chat_message.c.chat_id.in_(unresolved_chat_ids))
+                ):
+                    values = row._mapping
+                    chat_id = values['chat_id']
+                    prefix = f'{chat_id}-'
+                    message_id = values['id']
+                    if message_id and message_id.startswith(prefix):
+                        message_id = message_id[len(prefix) :]
+                    if not message_id:
+                        continue
+                    parent_id = values['parent_id']
+                    if parent_id and parent_id.startswith(prefix):
+                        parent_id = parent_id[len(prefix) :]
+                    messages_by_chat.setdefault(chat_id, {})[message_id] = {
+                        'parent_id': parent_id,
+                        'created_at': values['created_at'] or 0,
+                    }
+
+            for chat_id, messages in messages_by_chat.items():
+                parent_ids = {
+                    message['parent_id']
+                    for message in messages.values()
+                    if message.get('parent_id') in messages
+                }
+                leaf_ids = [message_id for message_id in messages if message_id not in parent_ids]
+                resolved[chat_id] = max(
+                    leaf_ids or list(messages),
+                    key=lambda message_id: messages[message_id].get('created_at') or 0,
+                )
+
+        for chat_id in batch_chat_ids:
+            if chat_id in resolved:
+                continue
+
+            messages = json_messages_by_chat.get(chat_id, {})
+            valid_candidate = next(
+                (candidate for candidate in candidates_by_chat[chat_id] if candidate in messages),
+                None,
+            )
+            if valid_candidate:
+                resolved[chat_id] = valid_candidate
+            elif messages:
+                parent_ids = {
+                    message['parent_id']
+                    for message in messages.values()
+                    if message.get('parent_id') in messages
+                }
+                leaf_ids = [message_id for message_id in messages if message_id not in parent_ids]
+                resolved[chat_id] = max(
+                    leaf_ids or list(messages),
+                    key=lambda message_id: messages[message_id].get('created_at') or 0,
+                )
+
+        updates = [
+            {'chat_id': chat_id, 'current_message_id': message_id}
+            for chat_id, message_id in resolved.items()
+            if message_id and message_id != current_by_chat.get(chat_id)
+        ]
+        if updates:
             conn.execute(
                 sa.update(chat)
-                .where(chat.c.id == chat_id)
-                .values(current_message_id=current_message_id)
+                .where(chat.c.id == sa.bindparam('update_chat_id'))
+                .values(current_message_id=sa.bindparam('update_current_message_id')),
+                [
+                    {
+                        'update_chat_id': row['chat_id'],
+                        'update_current_message_id': row['current_message_id'],
+                    }
+                    for row in updates
+                ],
             )
 
 

From 6ff1df326c76824f0706671b0974df4035cb453f Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 03:05:29 -0400
Subject: [PATCH 08/29] refac

---
 backend/open_webui/utils/asgi_middleware.py |  6 +-----
 backend/open_webui/utils/auth.py            | 18 +++++++++---------
 2 files changed, 10 insertions(+), 14 deletions(-)

diff --git a/backend/open_webui/utils/asgi_middleware.py b/backend/open_webui/utils/asgi_middleware.py
index 1ed539fefa..89154b6103 100644
--- a/backend/open_webui/utils/asgi_middleware.py
+++ b/backend/open_webui/utils/asgi_middleware.py
@@ -39,7 +39,6 @@ from fastapi.responses import JSONResponse, RedirectResponse
 from fastapi.security import HTTPAuthorizationCredentials
 from open_webui.env import CUSTOM_API_KEY_HEADER
 from open_webui.internal.db import ScopedSession
-from open_webui.models.config import Config
 from open_webui.utils.auth import get_http_authorization_cred
 from starlette.datastructures import MutableHeaders
 from starlette.requests import Request
@@ -138,9 +137,7 @@ class AuthTokenMiddleware:
     the middleware checks that instead and avoids the 401 short-circuit.
 
     Routes that depend on `get_verified_user` etc. read this state.
-    Also exposes `request.state.enable_api_keys` (snapshotted at request
-    entry from runtime config) and stamps an `X-Process-Time` response
-    header.
+    Also stamps an `X-Process-Time` response header.
     """
 
     def __init__(self, app: ASGIApp, *, fastapi_app) -> None:
@@ -166,7 +163,6 @@ class AuthTokenMiddleware:
                 token = HTTPAuthorizationCredentials(scheme='Bearer', credentials=api_key)
 
         request.state.token = token
-        request.state.enable_api_keys = await Config.get('auth.enable_api_keys')
 
         async def send_with_timing(message: Message) -> None:
             if message['type'] == 'http.response.start':
diff --git a/backend/open_webui/utils/auth.py b/backend/open_webui/utils/auth.py
index c8ff94bc09..7e5dc24a91 100644
--- a/backend/open_webui/utils/auth.py
+++ b/backend/open_webui/utils/auth.py
@@ -437,24 +437,24 @@ async def get_current_user_by_api_key(request, api_key: str):
             detail=ERROR_MESSAGES.INVALID_TOKEN,
         )
 
-    user_permissions = await Config.get('user.permissions')
-    enable_endpoint_restrictions = await Config.get('auth.api_key.endpoint_restrictions')
-    allowed_endpoints = await Config.get('auth.api_key.allowed_endpoints', '')
+    if not await Config.get('auth.enable_api_keys'):
+        raise HTTPException(status.HTTP_403_FORBIDDEN, detail=ERROR_MESSAGES.API_KEY_NOT_ALLOWED)
 
-    if not request.state.enable_api_keys or (
-        user.role != 'admin'
-        and not await has_permission(
+    if user.role != 'admin':
+        user_permissions = await Config.get('user.permissions')
+        if not await has_permission(
             user.id,
             'features.api_keys',
             user_permissions,
-        )
-    ):
-        raise HTTPException(status.HTTP_403_FORBIDDEN, detail=ERROR_MESSAGES.API_KEY_NOT_ALLOWED)
+        ):
+            raise HTTPException(status.HTTP_403_FORBIDDEN, detail=ERROR_MESSAGES.API_KEY_NOT_ALLOWED)
 
     # Enforce endpoint restrictions — checked here (not in middleware)
     # so it applies regardless of how the API key was transported
     # (Authorization header, cookie, x-api-key header, etc.).
+    enable_endpoint_restrictions = await Config.get('auth.api_key.endpoint_restrictions')
     if enable_endpoint_restrictions:
+        allowed_endpoints = await Config.get('auth.api_key.allowed_endpoints', '')
         allowed_paths = [path.strip() for path in str(allowed_endpoints).split(',') if path.strip()]
         request_path = request.scope['path']  # Use raw ASGI path — not spoofable via Host header (CVE-2026-48710)
         is_allowed = any(request_path == allowed or request_path.startswith(allowed + '/') for allowed in allowed_paths)

From 30415c925a18b1ea1c3f2739bd944dd939f020cf Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 03:11:38 -0400
Subject: [PATCH 09/29] refac

---
 pyproject.toml |  1 -
 uv.lock        | 38 --------------------------------------
 2 files changed, 39 deletions(-)

diff --git a/pyproject.toml b/pyproject.toml
index 468543e71c..74659791df 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -151,7 +151,6 @@ all = [
     "psycopg2-binary==2.9.12",
     "pgvector==0.4.2",
     "moto[s3]>=5.0.26",
-    "gcp-storage-emulator>=2024.8.3",
     "docker~=7.1.0",
     "pytest~=8.3.2",
     "pytest-docker~=3.2.5",
diff --git a/uv.lock b/uv.lock
index f68b6ffc5f..404c6ebacc 100644
--- a/uv.lock
+++ b/uv.lock
@@ -191,15 +191,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/ba/16/9826f089383c593cdfc4a6e5aca94d9e91ae1692c57af82c3b2aa5e810f7/anyio-4.14.0-py3-none-any.whl", hash = "sha256:dd9b7a2a9799ed6552fde617b2c5df02b7fdd7d88392fc48101e51bae46164d9", size = 123506 },
 ]
 
-[[package]]
-name = "appdirs"
-version = "1.4.4"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/d7/d8/05696357e0311f5b5c316d7b95f46c669dd9c15aaeecbb48c7d0aeb88c40/appdirs-1.4.4.tar.gz", hash = "sha256:7d5d0167b2b1ba821647616af46a749d1c653740dd0d2415100fe26e27afdf41", size = 13470 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/3b/00/2344469e2084fb287c2e0b57b72910309874c3245463acd6cf5e3db69324/appdirs-1.4.4-py2.py3-none-any.whl", hash = "sha256:a841dacd6b99318a741b166adb07e19ee71a274450e68237b4650ca1055ab128", size = 9566 },
-]
-
 [[package]]
 name = "apscheduler"
 version = "3.11.2"
@@ -1375,20 +1366,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/9a/9a/e35b4a917281c0b8419d4207f4334c8e8c5dbf4f3f5f9ada73958d937dcc/frozenlist-1.8.0-py3-none-any.whl", hash = "sha256:0c18a16eab41e82c295618a77502e17b195883241c563b00f0aa5106fc4eaa0d", size = 13409 },
 ]
 
-[[package]]
-name = "fs"
-version = "2.4.16"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "appdirs" },
-    { name = "setuptools" },
-    { name = "six" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/5d/a9/af5bfd5a92592c16cdae5c04f68187a309be8a146b528eac3c6e30edbad2/fs-2.4.16.tar.gz", hash = "sha256:ae97c7d51213f4b70b6a958292530289090de3a7e15841e108fbe144f069d313", size = 187441 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/b9/5c/a3d95dc1ec6cdeb032d789b552ecc76effa3557ea9186e1566df6aac18df/fs-2.4.16-py2.py3-none-any.whl", hash = "sha256:660064febbccda264ae0b6bace80a8d1be9e089e0a5eb2427b7d517f9a91545c", size = 135261 },
-]
-
 [[package]]
 name = "fsspec"
 version = "2025.3.0"
@@ -1415,19 +1392,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/ab/6e/81d47999aebc1b155f81eca4477a616a70f238a2549848c38983f3c22a82/ftfy-6.3.1-py3-none-any.whl", hash = "sha256:7c70eb532015cd2f9adb53f101fb6c7945988d023a085d127d1573dc49dd0083", size = 44821 },
 ]
 
-[[package]]
-name = "gcp-storage-emulator"
-version = "2024.8.3"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "fs" },
-    { name = "google-crc32c" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/17/c2/a0b0e1e54fdd9453603d90939faf652e2488b617c8752edc4ebcd89f1686/gcp_storage_emulator-2024.8.3.tar.gz", hash = "sha256:e5d45e5c23a0344c1c4c44b8f8c36f7e8975ca1fcc5134cab608b96ddccd9225", size = 24928 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/53/bf/b6c717bd7a5b59244388057d36789e72c28bfaa5e51f3a494563a4e3028e/gcp_storage_emulator-2024.8.3-py3-none-any.whl", hash = "sha256:1dc4ea56a0caf50fc6092898b9461d08b494824bfbbcca168e2af5da89c053ce", size = 19385 },
-]
-
 [[package]]
 name = "gitdb"
 version = "4.0.12"
@@ -3070,7 +3034,6 @@ all = [
     { name = "colbert-ai" },
     { name = "docker" },
     { name = "elasticsearch" },
-    { name = "gcp-storage-emulator" },
     { name = "moto", extra = ["s3"] },
     { name = "oracledb" },
     { name = "pgvector" },
@@ -3140,7 +3103,6 @@ requires-dist = [
     { name = "faster-whisper", specifier = "==1.2.1" },
     { name = "fpdf2", specifier = "==2.8.7" },
     { name = "ftfy", specifier = "==6.3.1" },
-    { name = "gcp-storage-emulator", marker = "extra == 'all'", specifier = ">=2024.8.3" },
     { name = "google-api-python-client", specifier = "==2.197.0" },
     { name = "google-auth-httplib2", specifier = "==0.4.0" },
     { name = "google-auth-oauthlib", specifier = "==1.4.0" },

From 9f00b62b3a005b030ffaf638fd1da4c27e3c0586 Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 03:22:51 -0400
Subject: [PATCH 10/29] refac

---
 backend/open_webui/routers/ollama.py | 58 ++++++++++++++++++++++++++++
 1 file changed, 58 insertions(+)

diff --git a/backend/open_webui/routers/ollama.py b/backend/open_webui/routers/ollama.py
index 60d95f176b..27a029c75d 100644
--- a/backend/open_webui/routers/ollama.py
+++ b/backend/open_webui/routers/ollama.py
@@ -1180,6 +1180,14 @@ class OpenAICompletionForm(BaseModel):
     model_config = ConfigDict(extra='allow')
 
 
+class OpenAIEmbeddingsForm(BaseModel):
+    """Payload for the OpenAI-compatible /v1/embeddings proxy."""
+
+    model: str
+    input: object
+    model_config = ConfigDict(extra='allow')
+
+
 @router.post('/v1/completions')
 @router.post('/v1/completions/{url_idx}')
 async def generate_openai_completion(
@@ -1235,6 +1243,56 @@ async def generate_openai_completion(
     )
 
 
+@router.post('/v1/embeddings')
+@router.post('/v1/embeddings/{url_idx}')
+async def generate_openai_embeddings(
+    request: Request,
+    form_data: dict,
+    url_idx: int | None = None,
+    user=Depends(get_verified_user),  # noqa: B008
+):
+    """Forward an embeddings request via the OpenAI-compatible proxy."""
+    if not await Config.get('ollama.enable'):
+        raise HTTPException(status_code=503, detail=ERROR_MESSAGES.OLLAMA_API_DISABLED)
+
+    metadata = form_data.pop('metadata', None)
+
+    try:
+        form_data = OpenAIEmbeddingsForm(**form_data)
+    except Exception as exc:
+        log.exception(exc)
+        raise HTTPException(status_code=400, detail=str(exc))
+
+    payload = {**form_data.model_dump(exclude_none=True)}
+    payload.pop('metadata', None)
+
+    model_id = form_data.model
+    model_info = await Models.get_model_by_id(model_id)
+    if model_info is not None:
+        if model_info.base_model_id:
+            payload['model'] = model_info.base_model_id
+        await check_model_access(user, model_info)
+    else:
+        await check_model_access(user, None)
+
+    url, url_idx = await get_ollama_url(request, payload['model'], url_idx, user)
+    api_config = resolve_api_config((await Config.get('ollama.api_configs', {})), url_idx, url)
+
+    prefix_id = api_config.get('prefix_id')
+    if prefix_id:
+        payload['model'] = payload['model'].replace(f'{prefix_id}.', '')
+
+    return await send_request(
+        f'{url}/v1/embeddings',
+        payload=json.dumps(payload),
+        key=get_api_key(url_idx, url, (await Config.get('ollama.api_configs', {}))),
+        user=user,
+        metadata=metadata,
+        api_config=api_config,
+        request=request,
+    )
+
+
 @router.post('/v1/chat/completions')
 @router.post('/v1/chat/completions/{url_idx}')
 async def generate_openai_chat_completion(

From 1b28b8a14444fb00cd3f7c2e963f40bd6f548059 Mon Sep 17 00:00:00 2001
From: joaoback <156559121+joaoback@users.noreply.github.com>
Date: Thu, 23 Jul 2026 04:23:47 -0300
Subject: [PATCH 11/29] i18n: add pt-BR translations for newly added UI items
 and consistency pass (#27324)

New **pt-BR** translations for items introduced in the latest releases, plus a consistency/quality pass across existing strings (grammar, tone, capitalization, pluralization). Placeholders and hotkeys preserved. No logic changes.
---
 src/lib/i18n/locales/pt-BR/translation.json | 1064 +++++++++----------
 1 file changed, 532 insertions(+), 532 deletions(-)

diff --git a/src/lib/i18n/locales/pt-BR/translation.json b/src/lib/i18n/locales/pt-BR/translation.json
index 744a916bd0..83f734e9cd 100644
--- a/src/lib/i18n/locales/pt-BR/translation.json
+++ b/src/lib/i18n/locales/pt-BR/translation.json
@@ -10,9 +10,9 @@
 	"{{COUNT}} Available Tools": "{{COUNT}} Ferramentas disponíveis",
 	"{{COUNT}} characters": "{{COUNT}} caracteres",
 	"{{COUNT}} extracted lines": "{{COUNT}} linhas extraídas",
-	"{{count}} files_one": "",
-	"{{count}} files_many": "",
-	"{{count}} files_other": "",
+	"{{count}} files_one": "{{count}} arquivo",
+	"{{count}} files_many": "{{count}} arquivos",
+	"{{count}} files_other": "{{count}} arquivos",
 	"{{COUNT}} files": "{{COUNT}} arquivos",
 	"{{count}} files selected. Only new and modified files will be uploaded. Deleted files will be removed. The folder structure will be mirrored. Continue?_one": "{{count}} arquivo selecionado. Apenas arquivos novos e modificados serão enviados. Arquivos excluídos serão removidos. A estrutura de pastas será espelhada. Continuar?",
 	"{{count}} files selected. Only new and modified files will be uploaded. Deleted files will be removed. The folder structure will be mirrored. Continue?_many": "{{count}} arquivos selecionados. Apenas arquivos novos e modificados serão enviados. Arquivos excluídos serão removidos. A estrutura de pastas será espelhada. Continuar?",
@@ -51,7 +51,7 @@
 	"{{user}}'s Chats": "Chats de {{user}}",
 	"{{webUIName}} Backend Required": "Backend {{webUIName}} necessário",
 	"*Prompt node ID(s) are required for image generation": "*Prompt node ID(s) são obrigatórios para gerar imagens",
-	"1 file": "",
+	"1 file": "1 arquivo",
 	"1 group": "1 grupo",
 	"1 hour before": "1 hora antes",
 	"1 Source": "1 Origem",
@@ -71,7 +71,7 @@
 	"Access": "Acesso",
 	"Access Control": "Controle de Acesso",
 	"Access Grants": "Concessões de Acesso",
-	"Access level": "",
+	"Access level": "Nível de acesso",
 	"Access List": "Lista de acesso",
 	"Access prohibited": "Acesso proibido",
 	"Access updated": "Acesso atualizado",
@@ -86,7 +86,7 @@
 	"Active": "Ativo",
 	"Active Users": "Usuários Ativos",
 	"Activity": "Atividade",
-	"Activity insights": "",
+	"Activity insights": "Insights de atividade",
 	"Add": "Adicionar",
 	"Add a model ID": "Adicione um ID de modelo",
 	"Add a preference, fact, or instruction about you": "Adicione uma preferência, fato ou instrução sobre você",
@@ -95,7 +95,7 @@
 	"Add a tag...": "Adicione uma tag...",
 	"Add Access": "Adicionar acesso",
 	"Add Arena Model": "Adicionar Modelo Arena",
-	"Add banner": "",
+	"Add banner": "Adicionar banner",
 	"Add Connection": "Adicionar Conexão",
 	"Add Content": "Adicionar Conteúdo",
 	"Add content here": "Adicionar conteúdo aqui",
@@ -104,7 +104,7 @@
 	"Add description": "Adicionar descrição",
 	"Add Details": "Adicionar detalhes",
 	"Add durable context for future chats": "Adicionar contexto durável para chats futuros",
-	"Add filenames, titles, sections, and snippets to improve lexical recall.": "",
+	"Add filenames, titles, sections, and snippets to improve lexical recall.": "Adicione nomes de arquivos, títulos, seções e trechos para melhorar a recuperação lexical.",
 	"Add Files": "Adicionar Arquivos",
 	"Add Image": "Adicionar imagem",
 	"Add Knowledge Connection": "Adicionar Conexão de Conhecimento",
@@ -112,9 +112,9 @@
 	"Add Member": "Adicionar membro",
 	"Add Members": "Adicionar membros",
 	"Add Memory": "Adicionar Memória",
-	"Add Notification Target": "",
-	"Add optional profile context visible where profiles are shown.": "",
-	"Add prompt suggestion": "",
+	"Add Notification Target": "Adicionar destino de notificação",
+	"Add optional profile context visible where profiles are shown.": "Adicione um contexto de perfil opcional, visível onde os perfis são exibidos.",
+	"Add prompt suggestion": "Adicionar sugestão de prompt",
 	"Add Reaction": "Adicionar reação",
 	"Add tag": "Adicionar tag",
 	"Add Tag": "Adicionar Tag",
@@ -129,27 +129,27 @@
 	"Add your Open Terminal URL and API key in Settings → Integrations.": "Adicione a URL do Open Terminal e a chave de API em Configurações → Integrações.",
 	"Additional Config": "Configuração adicional",
 	"Additional configuration options for marker. This should be a JSON string with key-value pairs. For example, '{\"key\": \"value\"}'. Supported keys include: disable_links, keep_pageheader_in_output, keep_pagefooter_in_output, filter_blank_pages, drop_repeated_text, layout_coverage_threshold, merge_threshold, height_tolerance, gap_threshold, image_threshold, min_line_length, level_count, default_level": "Opções de configuração adicionais para o marcador. Deve ser uma string JSON com pares chave-valor. Por exemplo, '{\"key\": \"value\"}'. As chaves suportadas incluem: disable_links, keep_pageheader_in_output, keep_pagefooter_in_output, filter_blank_pages, drop_repeated_text, layout_coverage_threshold, merge_threshold, height_tolerance, gap_threshold, image_threshold, min_line_length, level_count, default_level",
-	"Additional Docling parameters in JSON format.": "",
+	"Additional Docling parameters in JSON format.": "Parâmetros adicionais do Docling em formato JSON.",
 	"Additional feedback comments": "Comentários adicionais de feedback",
-	"Additional JSON headers sent to the external document loader.": "",
+	"Additional JSON headers sent to the external document loader.": "Cabeçalhos JSON adicionais enviados ao carregador de documentos externo.",
 	"Additional Parameters": "Parâmetros adicionais",
-	"Adjust how quickly spoken responses are played.": "",
+	"Adjust how quickly spoken responses are played.": "Ajuste a velocidade de reprodução das respostas faladas.",
 	"Adjusting these settings will apply changes universally to all users.": "Ajustar essas configurações aplicará mudanças para todos os usuários.",
 	"admin": "admin",
 	"Admin": "Admin",
 	"Admin Contact Email": "E-mail de contato do administrador",
-	"Admin Details": "",
+	"Admin Details": "Detalhes do administrador",
 	"Admin Panel": "Painel do Admin",
 	"Admin Roles": "Funções de Admin",
 	"Admin-managed service account": "Conta de serviço gerenciada pelo admin",
 	"Admins have access to all tools at all times; users need tools assigned per model in the workspace.": "Os administradores têm acesso a todas as ferramentas o tempo todo; os usuários precisam de ferramentas atribuídas, por modelo, no workspace.",
 	"Advanced": "Avançado",
-	"Advanced MinerU parsing parameters in JSON format.": "",
+	"Advanced MinerU parsing parameters in JSON format.": "Parâmetros avançados de análise do MinerU em formato JSON.",
 	"Advanced Parameters": "Parâmetros Avançados",
 	"Advanced Params": "Parâmetros Avançados",
-	"After changing the embedding model, reindex the knowledge base for changes to take effect.": "",
+	"After changing the embedding model, reindex the knowledge base for changes to take effect.": "Após alterar o modelo de embedding, reindexe a base de conhecimento para que as alterações entrem em vigor.",
 	"All": "Tudo",
-	"All Chats": "",
+	"All Chats": "Todos os chats",
 	"All chats have been unarchived.": "Todos os chats foram desarquivados.",
 	"All day": "O dia todo",
 	"All events": "Todos os eventos",
@@ -161,9 +161,9 @@
 	"All time": "Todo o período",
 	"All Users": "Todos os usuários",
 	"All users and system events": "Todos os usuários e eventos do sistema",
-	"Allow artifacts to access same-origin browser APIs inside the sandbox.": "",
-	"Allow automatic names for new chats.": "",
-	"Allow browser notifications for completed responses.": "",
+	"Allow artifacts to access same-origin browser APIs inside the sandbox.": "Permita que artefatos acessem APIs do navegador da mesma origem dentro do sandbox.",
+	"Allow automatic names for new chats.": "Permita nomes automáticos para novos chats.",
+	"Allow browser notifications for completed responses.": "Permita notificações do navegador para respostas concluídas.",
 	"Allow Call": "Permitir chamada",
 	"Allow Chat Controls": "Permitir Controles de Chats",
 	"Allow Chat Delete": "Permitir Exclusão de Chats",
@@ -175,41 +175,41 @@
 	"Allow Chat System Prompt": "Permitir Prompt do Sistema no Chat",
 	"Allow Chat Valves": "Permitir Configurações de Chat",
 	"Allow Continue Response": "Permitir Resposta Contínua",
-	"Allow delegated sub-agents to keep running while the parent chat continues.": "",
+	"Allow delegated sub-agents to keep running while the parent chat continues.": "Permita que subagentes delegados continuem em execução enquanto o chat principal prossegue.",
 	"Allow Delete Messages": "Permitir Exclusão de Mensagens",
 	"Allow File Upload": "Permitir Envio de arquivos",
-	"Allow forms inside sandboxed artifact iframes.": "",
-	"Allow Google Drive as a document source.": "",
-	"Allow models to run generated code and return execution results.": "",
-	"Allow models to use the code interpreter tool during chats.": "",
+	"Allow forms inside sandboxed artifact iframes.": "Permita formulários dentro de iframes de artefatos em sandbox.",
+	"Allow Google Drive as a document source.": "Permita o Google Drive como fonte de documentos.",
+	"Allow models to run generated code and return execution results.": "Permita que os modelos executem o código gerado e retornem os resultados da execução.",
+	"Allow models to use the code interpreter tool during chats.": "Permita que os modelos usem a ferramenta de interpretação de código durante os chats.",
 	"Allow Multiple Models in Chat": "Permitir Vários Modelos no Chat",
-	"Allow new users to create accounts.": "",
+	"Allow new users to create accounts.": "Permita que novos usuários criem contas.",
 	"Allow non-local voices": "Permitir vozes não locais",
-	"Allow OneDrive as a document source.": "",
+	"Allow OneDrive as a document source.": "Permita o OneDrive como fonte de documentos.",
 	"Allow public write access": "Permitir acesso público de escrita",
 	"Allow Rate Response": "Permitir Avaliar Resposta",
 	"Allow Regenerate Response": "Permitir Regenerar Resposta",
-	"Allow Sharing With Groups": "",
+	"Allow Sharing With Groups": "Permitir compartilhamento com grupos",
 	"Allow Sharing With Users": "Permitir Compartilhamento com Usuários",
 	"Allow Speech to Text": "Permitir Fala para Texto",
 	"Allow Temporary Chat": "Permitir Conversa Temporária",
 	"Allow Text to Speech": "Permitir Texto para Fala",
-	"Allow the AI to delegate tasks to sub-agents. Each sub-agent creates a real chat with full tool access. Uses additional LLM calls.": "",
+	"Allow the AI to delegate tasks to sub-agents. Each sub-agent creates a real chat with full tool access. Uses additional LLM calls.": "Permita que a IA delegue tarefas a subagentes. Cada subagente cria um chat real com acesso completo às ferramentas. Usa chamadas adicionais ao LLM.",
 	"Allow User Location": "Permitir Localização do Usuário",
-	"Allow users to access calendar features.": "",
-	"Allow users to authenticate with an LDAP directory.": "",
-	"Allow users to configure webhooks from their account.": "",
-	"Allow users to create accounts through OAuth.": "",
-	"Allow users to create and manage notes.": "",
-	"Allow users to create and run automations.": "",
-	"Allow users to create API keys for programmatic access.": "",
-	"Allow users to edit existing images.": "",
-	"Allow users to generate images from prompts.": "",
-	"Allow users to organize chats into folders.": "",
-	"Allow users to save memories for more personalized responses.": "",
-	"Allow users to search the web from chats.": "",
-	"Allow users to share chats with the Open WebUI community.": "",
-	"Allow users to use channels for shared conversations.": "",
+	"Allow users to access calendar features.": "Permita que os usuários acessem os recursos de calendário.",
+	"Allow users to authenticate with an LDAP directory.": "Permita que os usuários se autentiquem com um diretório LDAP.",
+	"Allow users to configure webhooks from their account.": "Permita que os usuários configurem webhooks em suas contas.",
+	"Allow users to create accounts through OAuth.": "Permita que os usuários criem contas por meio do OAuth.",
+	"Allow users to create and manage notes.": "Permita que os usuários criem e gerenciem notas.",
+	"Allow users to create and run automations.": "Permita que os usuários criem e executem automações.",
+	"Allow users to create API keys for programmatic access.": "Permita que os usuários criem chaves de API para acesso programático.",
+	"Allow users to edit existing images.": "Permita que os usuários editem imagens existentes.",
+	"Allow users to generate images from prompts.": "Permita que os usuários gerem imagens a partir de prompts.",
+	"Allow users to organize chats into folders.": "Permita que os usuários organizem chats em pastas.",
+	"Allow users to save memories for more personalized responses.": "Permita que os usuários salvem memórias para obter respostas mais personalizadas.",
+	"Allow users to search the web from chats.": "Permita que os usuários pesquisem na web a partir dos chats.",
+	"Allow users to share chats with the Open WebUI community.": "Permita que os usuários compartilhem chats com a comunidade Open WebUI.",
+	"Allow users to use channels for shared conversations.": "Permita que os usuários usem canais para conversas compartilhadas.",
 	"Allow Voice Interruption in Call": "Permitir Interrupção de Voz na Chamada",
 	"Allow Web Upload": "Permitir Upload da Web",
 	"Allowed Domains": "Domínios Permitidos",
@@ -217,7 +217,7 @@
 	"Allowed File Extensions": "Extensões de arquivo permitidas",
 	"Allowed Roles": "Funções Permitidas",
 	"Already have an account?": "Já possui uma conta?",
-	"Also bound to {{action}}": "",
+	"Also bound to {{action}}": "Também vinculado a {{action}}",
 	"Alternative to the top_p, and aims to ensure a balance of quality and variety. The parameter p represents the minimum probability for a token to be considered, relative to the probability of the most likely token. For example, with p=0.05 and the most likely token having a probability of 0.9, logits with a value less than 0.045 are filtered out.": "Alternativa ao top_p, visando garantir um equilíbrio entre qualidade e variedade. O parâmetro p representa a probabilidade mínima de um token ser considerado, em relação à probabilidade do token mais provável. Por exemplo, com p = 0,05 e o token mais provável tendo uma probabilidade de 0,9, logits com valor inferior a 0,045 são filtrados.",
 	"Always": "Sempre",
 	"Always Collapse Code Blocks": "Sempre recolher blocos de código",
@@ -231,35 +231,35 @@
 	"and create a new shared link.": "e criar um novo link compartilhado.",
 	"Android": "Android",
 	"Anyone": "Qualquer um",
-	"API Auth String": "",
+	"API Auth String": "String de autenticação da API",
 	"API Base URL": "URL Base da API",
 	"API Key": "Chave API",
 	"API Key / Token": "Chave API / Token",
 	"API Key created.": "Chave API criada.",
 	"API Key Endpoint Restrictions": "Restrições de endpoint de chave de API",
-	"API key for embedding requests.": "",
-	"API key sent to Docling.": "",
-	"API key sent to Mistral OCR.": "",
-	"API key sent to the external document loader.": "",
-	"API key sent to the external reranker.": "",
-	"API key used for MinerU cloud mode.": "",
-	"API key used to authenticate with Datalab Marker.": "",
+	"API key for embedding requests.": "Chave de API para solicitações de embedding.",
+	"API key sent to Docling.": "Chave de API enviada ao Docling.",
+	"API key sent to Mistral OCR.": "Chave de API enviada ao Mistral OCR.",
+	"API key sent to the external document loader.": "Chave de API enviada ao carregador de documentos externo.",
+	"API key sent to the external reranker.": "Chave de API enviada ao reranqueador externo.",
+	"API key used for MinerU cloud mode.": "Chave de API usada no modo cloud do MinerU.",
+	"API key used to authenticate with Datalab Marker.": "Chave de API usada para autenticação com o Datalab Marker.",
 	"API keys": "Chaves API",
 	"API Keys": "Chaves API",
 	"API Mode": "Modo API",
 	"API Timeout": "Tempo limite da API",
-	"API Token": "",
-	"API token sent to PaddleOCR-vl.": "",
+	"API Token": "Token da API",
+	"API token sent to PaddleOCR-vl.": "Token da API enviado ao PaddleOCR-vl.",
 	"API Type": "Tipo de API",
-	"API URL": "",
+	"API URL": "URL da API",
 	"API Version": "Versão da API",
 	"API Version is required": "Versão da API é obrigatória",
-	"Append a watermark to assistant responses when configured.": "",
+	"Append a watermark to assistant responses when configured.": "Adicione uma marca d'água às respostas do assistente quando configurada.",
 	"Application DN": "DN da Aplicação",
 	"Application DN Password": "Senha da aplicação DN",
 	"applies to all users with the \"user\" role": "Aplica-se a todos com a função de \"usuário\"",
-	"Apply image compression to channel uploads too.": "",
-	"Apply voice-specific instructions while voice mode is active.": "",
+	"Apply image compression to channel uploads too.": "Aplique a compressão de imagens também aos uploads em canais.",
+	"Apply voice-specific instructions while voice mode is active.": "Aplique instruções específicas de voz enquanto o modo de voz estiver ativo.",
 	"April": "Abril",
 	"Archive": "Arquivar",
 	"Archive All": "Arquivar tudo",
@@ -290,9 +290,9 @@
 	"Ask a test question": "Fazer uma pergunta de teste",
 	"Ask this knowledge source a test question": "Fazer uma pergunta de teste a esta fonte de conhecimento",
 	"Assistant": "Assistente",
-	"Assistant messages": "",
+	"Assistant messages": "Mensagens do assistente",
 	"Async Embedding Processing": "Processamento de Embedding assíncrono",
-	"at": "",
+	"at": "às",
 	"At time of event": "No horário do evento",
 	"Attach File From Knowledge": "Anexar arquivo da base de conhecimento",
 	"Attach Files": "Anexar arquivos",
@@ -319,14 +319,14 @@
 	"Autocomplete Generation Input Max Length": "Comprimento máximo de entrada de geração de preenchimento automático",
 	"Autocomplete Generation Prompt": "Prompt de Geração de Preenchimento Automático",
 	"Automatic": "Automático",
-	"Automatic Delivery": "",
-	"Automatic Events": "",
+	"Automatic Delivery": "Entrega automática",
+	"Automatic Events": "Eventos automáticos",
 	"Automatic1111": "Automatic1111",
 	"AUTOMATIC1111 Base URL is required.": "URL Base AUTOMATIC1111 é necessária.",
 	"Automatically inject system tools in native function calling mode (e.g., timestamps, memory, chat history, notes, etc.)": "Injetar automaticamente ferramentas do sistema no modo de chamada de função nativa (por exemplo, carimbos de data/hora, memória, histórico de chat, notas, etc.)",
 	"Automation": "Automação",
 	"Automation created": "Automação criada",
-	"Automation Menu": "",
+	"Automation Menu": "Menu de automação",
 	"Automation title": "Título de automação",
 	"Automation triggered": "Automação acionada",
 	"Automation updated": "Automação atualizada",
@@ -337,31 +337,31 @@
 	"Available Tools": "Ferramentas disponíveis",
 	"Available variables": "Variáveis disponíveis",
 	"available!": "disponível!",
-	"Average messages per active day": "",
-	"Average tokens per chat": "",
+	"Average messages per active day": "Média de mensagens por dia ativo",
+	"Average tokens per chat": "Média de tokens por chat",
 	"Away": "Ausente",
 	"Awful": "Horrível",
 	"Azure AI Speech": "Fala de IA do Azure",
 	"Azure OpenAI": "Azure OpenAI",
-	"Azure OpenAI API key.": "",
-	"Azure OpenAI API version.": "",
-	"Azure OpenAI endpoint used for embeddings.": "",
+	"Azure OpenAI API key.": "Chave de API do Azure OpenAI.",
+	"Azure OpenAI API version.": "Versão da API do Azure OpenAI.",
+	"Azure OpenAI endpoint used for embeddings.": "Endpoint do Azure OpenAI usado para embeddings.",
 	"Azure Region": "Região Azure",
 	"Back": "Voltar",
-	"Background sub-agent finished": "",
-	"background sub-agents": "",
+	"Background sub-agent finished": "Subagente em segundo plano concluído",
+	"background sub-agents": "subagentes em segundo plano",
 	"Bad Response": "Resposta Ruim",
-	"Balance semantic and lexical weighting for hybrid search.": "",
+	"Balance semantic and lexical weighting for hybrid search.": "Equilibre os pesos semântico e lexical na pesquisa híbrida.",
 	"Banners": "Banners",
-	"Base DN used when searching for users.": "",
+	"Base DN used when searching for users.": "DN base usado ao pesquisar usuários.",
 	"Base Model (From)": "Modelo Base (De)",
 	"Base Model is required.": "Modelo Base é obrigatório.",
 	"Base Model List Cache speeds up access by fetching base models only at startup or on settings save—faster, but may not show recent base model changes.": "O cache da lista de modelos base acelera o acesso buscando modelos base somente na inicialização ou ao salvar as configurações — mais rápido, mas pode não mostrar alterações recentes no modelo base.",
-	"Base URL": "",
+	"Base URL": "URL base",
 	"Bearer": "Bearer",
 	"before": "antes",
 	"Being lazy": "Sendo preguiçoso",
-	"Bind DN used for directory search.": "",
+	"Bind DN used for directory search.": "DN de vinculação usado na pesquisa no diretório.",
 	"Bing": "Bing",
 	"Bing Search V7 Endpoint": "Endpoint do Bing Search V7",
 	"Bing Search V7 Subscription Key": "Chave de assinatura do Bing Search V7",
@@ -376,7 +376,7 @@
 	"Brave Search API Key": "Chave API do Brave Search",
 	"Break down complex requests into trackable steps": "Divida solicitações complexas em etapas rastreáveis.",
 	"Browse and query knowledge bases": "Navegue e consulte bases de conhecimento.",
-	"Browser Notifications": "",
+	"Browser Notifications": "Notificações do navegador",
 	"Builtin Tools": "Ferramentas integradas",
 	"Bullet List": "Lista com marcadores",
 	"Button ID": "ID do botão",
@@ -384,7 +384,7 @@
 	"Button Prompt": "Prompt do botão",
 	"by {{name}}": "por {{name}}",
 	"By {{name}}": "Por {{name}}",
-	"By supporting the project through sponsorship or an enterprise license, you help us stay independent, ship new features faster, improve stability, and grow Open WebUI for the long haul.": "",
+	"By supporting the project through sponsorship or an enterprise license, you help us stay independent, ship new features faster, improve stability, and grow Open WebUI for the long haul.": "Ao apoiar o projeto por meio de patrocínio ou de uma licença empresarial, você nos ajuda a manter a independência, lançar novos recursos mais rapidamente, melhorar a estabilidade e desenvolver o Open WebUI a longo prazo.",
 	"Bypass Embedding and Retrieval": "Ignorar Embedding e Recuperação",
 	"Bypass Web Loader": "Ignorar carregador da Web",
 	"Cache Base Model List": "Lista de modelos base de cache",
@@ -396,7 +396,7 @@
 	"Calendars Public Sharing": "Compartilhamento Público de Calendários",
 	"Call": "Chamada",
 	"Call feature is not supported when using Web STT engine": "O recurso de chamada não é suportado ao usar o mecanismo Web STT",
-	"Callback URI registered with the provider.": "",
+	"Callback URI registered with the provider.": "URI de callback registrada no provedor.",
 	"Camera": "Câmera",
 	"Cancel": "Cancelar",
 	"Cancel download of {{model}}": "Cancelar download de {{model}}",
@@ -405,7 +405,7 @@
 	"Capabilities": "Capacidades",
 	"Capture": "Capturar",
 	"Capture Audio": "Capturar Audio",
-	"Certificate file used for TLS verification.": "",
+	"Certificate file used for TLS verification.": "Arquivo de certificado usado para verificação TLS.",
 	"Certificate Path": "Caminho do Certificado",
 	"Change folder icon": "Alterar ícone da pasta",
 	"Change Password": "Alterar Senha",
@@ -420,7 +420,7 @@
 	"Channels": "Canais",
 	"Character": "Caractere",
 	"Chat": "Chat",
-	"Chat actions": "",
+	"Chat actions": "Ações do chat",
 	"Chat archived.": "Chat arquivado.",
 	"Chat Background Image": "Imagem de Fundo do Chat",
 	"Chat Bubble UI": "Interface de Bolha de Chat",
@@ -429,11 +429,11 @@
 	"Chat deleted.": "Chat excluído.",
 	"Chat Direction": "Direção do Chat",
 	"Chat exported successfully": "Chat exportado com sucesso",
-	"Chat forked": "",
-	"Chat history": "",
+	"Chat forked": "Chat bifurcado",
+	"Chat history": "Histórico do chat",
 	"Chat History": "Histórico de chat",
 	"Chat ID": "ID do Chat",
-	"Chat is too short to compact": "",
+	"Chat is too short to compact": "O chat é muito curto para compactar",
 	"Chat moved successfully": "Chat movido com sucesso",
 	"Chat Permissions": "Permissões de Chat",
 	"Chat Tags Auto-Generation": "Tags de Chat Geradas Automaticamente",
@@ -445,47 +445,47 @@
 	"Check for updates": "Verificar atualizações",
 	"Checking for updates...": "Verificando atualizações...",
 	"Choose a model before saving...": "Escolha um modelo antes de salvar...",
-	"Choose automatic, left-to-right, or right-to-left text flow.": "",
-	"Choose fallback models for background tasks. Current Model follows the active chat model.": "",
-	"Choose how extracted text is split before indexing.": "",
-	"Choose how uploaded documents are parsed before indexing.": "",
-	"Choose how web result pages are fetched and read.": "",
-	"Choose the browser voice used for speech output.": "",
-	"Choose the color theme used by the interface.": "",
-	"Choose the configured text-to-speech service voice.": "",
-	"Choose the engine used to read assistant responses aloud.": "",
-	"Choose the engine used to transcribe voice input.": "",
-	"Choose the gender value stored on your profile.": "",
-	"Choose the Kokoro.js voice used for speech output.": "",
-	"Choose the language used for interface text.": "",
-	"Choose the local or cloud MinerU API mode.": "",
-	"Choose the provider used for image edits.": "",
-	"Choose the provider used for image generation.": "",
-	"Choose the provider used for web search queries.": "",
-	"Choose the runtime used by the code interpreter tool.": "",
-	"Choose the runtime used for generated code blocks.": "",
-	"Choose the speech provider used for assistant audio output.": "",
-	"Choose the transcription provider used for audio input.": "",
-	"Choose whether Enter sends immediately or uses Ctrl+Enter.": "",
-	"Choose whether the app opens to the default home or chat view.": "",
+	"Choose automatic, left-to-right, or right-to-left text flow.": "Escolha o fluxo de texto automático, da esquerda para a direita ou da direita para a esquerda.",
+	"Choose fallback models for background tasks. Current Model follows the active chat model.": "Escolha modelos alternativos para tarefas em segundo plano. Modelo Atual acompanha o modelo ativo no chat.",
+	"Choose how extracted text is split before indexing.": "Escolha como o texto extraído é dividido antes da indexação.",
+	"Choose how uploaded documents are parsed before indexing.": "Escolha como os documentos enviados são analisados antes da indexação.",
+	"Choose how web result pages are fetched and read.": "Escolha como as páginas de resultados da web são obtidas e lidas.",
+	"Choose the browser voice used for speech output.": "Escolha a voz do navegador usada na saída de fala.",
+	"Choose the color theme used by the interface.": "Escolha o tema de cores usado pela interface.",
+	"Choose the configured text-to-speech service voice.": "Escolha a voz configurada do serviço de texto para fala.",
+	"Choose the engine used to read assistant responses aloud.": "Escolha o mecanismo usado para ler as respostas do assistente em voz alta.",
+	"Choose the engine used to transcribe voice input.": "Escolha o mecanismo usado para transcrever a entrada de voz.",
+	"Choose the gender value stored on your profile.": "Escolha o valor de gênero armazenado em seu perfil.",
+	"Choose the Kokoro.js voice used for speech output.": "Escolha a voz do Kokoro.js usada na saída de fala.",
+	"Choose the language used for interface text.": "Escolha o idioma usado no texto da interface.",
+	"Choose the local or cloud MinerU API mode.": "Escolha o modo de API local ou cloud do MinerU.",
+	"Choose the provider used for image edits.": "Escolha o provedor usado para editar imagens.",
+	"Choose the provider used for image generation.": "Escolha o provedor usado para gerar imagens.",
+	"Choose the provider used for web search queries.": "Escolha o provedor usado nas consultas de pesquisa na web.",
+	"Choose the runtime used by the code interpreter tool.": "Escolha o runtime usado pela ferramenta de interpretação de código.",
+	"Choose the runtime used for generated code blocks.": "Escolha o runtime usado nos blocos de código gerados.",
+	"Choose the speech provider used for assistant audio output.": "Escolha o provedor de fala usado na saída de áudio do assistente.",
+	"Choose the transcription provider used for audio input.": "Escolha o provedor de transcrição usado na entrada de áudio.",
+	"Choose whether Enter sends immediately or uses Ctrl+Enter.": "Escolha se Enter envia imediatamente ou se Ctrl+Enter deve ser usado.",
+	"Choose whether the app opens to the default home or chat view.": "Escolha se o aplicativo abre na página inicial padrão ou na visualização do chat.",
 	"Chunk Min Size Target": "Tamanho mínimo alvo do bloco",
 	"Chunk Overlap": "Sobreposição de Chunk",
 	"Chunk Size": "Tamanho do Chunk",
 	"Ciphers": "Cifras",
 	"Citation": "Citação",
 	"Citations": "Citações",
-	"Claim containing provider groups.": "",
-	"Claim containing provider roles.": "",
-	"Claim used as the display name.": "",
-	"Claim used as the profile picture URL.": "",
-	"Claim used as the stable user identifier.": "",
-	"Claim used as the user email address.": "",
-	"Clear filters to reorder": "",
+	"Claim containing provider groups.": "Claim que contém os grupos do provedor.",
+	"Claim containing provider roles.": "Claim que contém as funções do provedor.",
+	"Claim used as the display name.": "Claim usada como nome de exibição.",
+	"Claim used as the profile picture URL.": "Claim usada como URL da imagem de perfil.",
+	"Claim used as the stable user identifier.": "Claim usada como identificador estável do usuário.",
+	"Claim used as the user email address.": "Claim usada como endereço de e-mail do usuário.",
+	"Clear filters to reorder": "Limpe os filtros para reordenar",
 	"Clear memory": "Limpar memória",
 	"Clear Memory": "Limpar Memória",
 	"Clear search": "Limpar pesquisa",
 	"Clear status": "Limpar status",
-	"Clear vector storage and knowledge indexing data.": "",
+	"Clear vector storage and knowledge indexing data.": "Limpe o armazenamento vetorial e os dados de indexação de conhecimento.",
 	"click here": "Clique aqui",
 	"Click here for filter guides.": "Clique aqui para obter instruções de filtros.",
 	"Click here for help.": "Clique aqui para obter ajuda.",
@@ -500,7 +500,7 @@
 	"Click on the user role button to change a user's role.": "Clique no botão de perfil do usuário para alterar o perfil de um usuário.",
 	"Click to connect": "Clique para conectar",
 	"Click to copy ID": "Clique para copiar o ID",
-	"Click to rebind": "",
+	"Click to rebind": "Clique para vincular novamente",
 	"Client ID": "ID do Cliente",
 	"Client Secret": "Chave Secreta do Cliente",
 	"Clipboard write permission denied. Please check your browser settings to grant the necessary access.": "Permissão de escrita na área de transferência negada. Verifique as configurações do seu navegador para conceder o acesso necessário.",
@@ -530,37 +530,37 @@
 	"Code Interpreter Prompt Template": "Modelo de Prompt do Interpretador de Código",
 	"Collaboration channel where people join as members": "Canal de colaboração onde as pessoas se juntam como membros.",
 	"Collapse": "Recolher",
-	"Collapse code blocks by default.": "",
+	"Collapse code blocks by default.": "Recolha os blocos de código por padrão.",
 	"Collection": "Coleção",
 	"Collection Field": "Campo de Coleção",
 	"Collections": "Coleções",
 	"Color": "Cor",
-	"Combine semantic and keyword retrieval.": "",
+	"Combine semantic and keyword retrieval.": "Combine a recuperação semântica e por palavras-chave.",
 	"ComfyUI": "ComfyUI",
 	"ComfyUI Base URL is required.": "URL Base do ComfyUI é necessária.",
 	"ComfyUI Workflow": "Fluxo de trabalho ComfyUI",
 	"ComfyUI Workflow Nodes": "Nós do fluxo de trabalho ComfyUI",
 	"Comma separated Node Ids (e.g. 1 or 1,2)": "IDs de Nodes separados por vírgula (por exemplo, 1 ou 1,2)",
-	"Comma-separated API paths that API keys can access.": "",
-	"Comma-separated audio or video MIME types accepted for upload.": "",
-	"Comma-separated extensions MinerU should handle.": "",
+	"Comma-separated API paths that API keys can access.": "Caminhos de API separados por vírgulas que as chaves de API podem acessar.",
+	"Comma-separated audio or video MIME types accepted for upload.": "Tipos MIME de áudio ou vídeo aceitos para upload, separados por vírgulas.",
+	"Comma-separated extensions MinerU should handle.": "Extensões que o MinerU deve processar, separadas por vírgulas.",
 	"Comma-separated group names": "Nomes de grupos separados por vírgula",
-	"Comma-separated upload extensions. Leave empty for all file types.": "",
+	"Comma-separated upload extensions. Leave empty for all file types.": "Extensões de upload separadas por vírgulas. Deixe vazio para permitir todos os tipos de arquivo.",
 	"command": "comando",
 	"Command": "Comando",
-	"Commands": "",
+	"Commands": "Comandos",
 	"Comment": "Comentário",
 	"Commit Message": "Mensagem de Commit",
-	"Community": "",
+	"Community": "Comunidade",
 	"Community Reviews": "Avaliações da comunidade",
-	"Community Sharing": "",
+	"Community Sharing": "Compartilhamento com a comunidade",
 	"Compacting context": "Compactando contexto",
-	"Compacting context...": "",
-	"Compare": "",
+	"Compacting context...": "Compactando contexto...",
+	"Compare": "Comparar",
 	"Comparing with knowledge base...": "Comparando com a base de conhecimento...",
 	"Completions": "Completions",
 	"Compress Images in Channels": "Comprimir imagens em canais",
-	"Compress uploaded images before sending or storage.": "",
+	"Compress uploaded images before sending or storage.": "Comprima as imagens enviadas antes do envio ou armazenamento.",
 	"Computing checksums ({{count}} files)_one": "Calculando checksums ({{count}} arquivo)",
 	"Computing checksums ({{count}} files)_many": "Calculando checksums ({{count}} arquivos)",
 	"Computing checksums ({{count}} files)_other": "Calculando checksums ({{count}} arquivos)",
@@ -574,14 +574,14 @@
 	"Confirm your action": "Confirme sua ação",
 	"Confirm your new password": "Confirme sua nova senha",
 	"Confirm Your Password": "Confirme sua senha",
-	"Connect code execution to a Jupyter server endpoint.": "",
-	"Connect code interpreter to a Jupyter server endpoint.": "",
-	"Connect to a stable-diffusion-webui server running with the `--api` flag.": "",
+	"Connect code execution to a Jupyter server endpoint.": "Conecte a execução de código a um endpoint de servidor Jupyter.",
+	"Connect code interpreter to a Jupyter server endpoint.": "Conecte o interpretador de código a um endpoint de servidor Jupyter.",
+	"Connect to a stable-diffusion-webui server running with the `--api` flag.": "Conecte-se a um servidor stable-diffusion-webui em execução com a flag `--api`.",
 	"Connect to an AI provider to start chatting": "Conecte-se a um provedor de IA para começar a conversar",
 	"Connect to Open Terminal instances to browse files and use them as always-on tools. Only one can be active at a time.": "Conecte-se a instâncias do Open Terminal para navegar em arquivos e usá-las como ferramentas sempre ativas. Apenas uma pode estar ativa por vez.",
 	"Connect to Open Terminal instances. All users will have access to file browsing and terminal tools through these servers.": "Conecte-se a instâncias do Open Terminal. Todos os usuários terão acesso à navegação de arquivos e ferramentas de terminal por meio desses servidores.",
-	"Connect to the ComfyUI server used for generation.": "",
-	"Connect to the ComfyUI server used for image edits.": "",
+	"Connect to the ComfyUI server used for generation.": "Conecte-se ao servidor ComfyUI usado para geração.",
+	"Connect to the ComfyUI server used for image edits.": "Conecte-se ao servidor ComfyUI usado para edição de imagens.",
 	"Connect to your own OpenAI compatible API endpoints.": "Conecte-se aos seus próprios endpoints de API compatíveis com OpenAI.",
 	"Connect to your own OpenAPI compatible external tool servers.": "Conecte-se aos seus próprios servidores de ferramentas externas compatíveis com OpenAPI.",
 	"Connected": "Conectado",
@@ -596,7 +596,7 @@
 	"Constrains effort on reasoning for reasoning models. Only applicable to reasoning models from specific providers that support reasoning effort.": "Restringe o esforço de raciocínio para modelos de raciocínio. Aplicável somente a modelos de raciocínio de provedores específicos que suportam o esforço de raciocínio.",
 	"Contact Admin for WebUI Access": "Contate o Administrador para verificar seu acesso",
 	"Content": "Conteúdo",
-	"Content Extraction": "",
+	"Content Extraction": "Extração de conteúdo",
 	"Content Extraction Engine": "Mecanismo de Extração de Conteúdo",
 	"Content Field": "Campo de Conteúdo",
 	"Content lengths (character counts only)": "Extensão do conteúdo (apenas em caracteres)",
@@ -604,7 +604,7 @@
 	"Context compacted": "Contexto compactado",
 	"Context Compaction": "Compactação de Contexto",
 	"Context compaction failed": "Falha na compactação de contexto",
-	"Context compaction is disabled": "",
+	"Context compaction is disabled": "A compactação de contexto está desativada",
 	"Context Compaction Prompt": "Prompt de Compactação de Contexto",
 	"Context Compaction Threshold": "Limite de Compactação de Contexto",
 	"Context Tokens": "Tokens de Contexto",
@@ -614,14 +614,14 @@
 	"Continue with Email": "Continuar com Email",
 	"Continue with LDAP": "Continuar com LDAP",
 	"Control how message text is split for TTS requests. 'Punctuation' splits into sentences, 'paragraphs' splits into paragraphs, and 'none' keeps the message as a single string.": "Controla como o texto da mensagem é dividido para solicitações TTS. 'Pontuação' divide em frases, 'Parágrafos' divide em parágrafos e 'Nenhum' mantém a mensagem como um texto único.",
-	"Control result volume and parallel search requests.": "",
+	"Control result volume and parallel search requests.": "Controle o volume de resultados e as solicitações de pesquisa paralelas.",
 	"Control the repetition of token sequences in the generated text. A higher value (e.g., 1.5) will penalize repetitions more strongly, while a lower value (e.g., 1.1) will be more lenient. At 1, it is disabled.": "Controla a repetição de sequências de tokens no texto gerado. Um valor mais alto (por exemplo, 1,5) penalizará as repetições com mais rigor, enquanto um valor mais baixo (por exemplo, 1,1) será mais tolerante. Em 1, ele está desabilitado.",
 	"Controls": "Controles",
-	"Controls how chat tags are inferred.": "",
-	"Controls how older messages are rewritten into a running summary.": "",
+	"Controls how chat tags are inferred.": "Controla como as tags do chat são inferidas.",
+	"Controls how older messages are rewritten into a running summary.": "Controla como as mensagens antigas são reescritas em um resumo contínuo.",
 	"Controls the balance between coherence and diversity of the output. A lower value will result in more focused and coherent text.": "Controla o equilíbrio entre coerência e diversidade da saída. Um valor menor resultará em um texto mais focado e coerente.",
 	"Conversation saved successfully": "Conversa salva com sucesso",
-	"Convert long pasted text into a file attachment.": "",
+	"Convert long pasted text into a file attachment.": "Converta textos longos colados em um arquivo anexado.",
 	"Copied": "Copiado",
 	"Copied link to clipboard": "Link copiado para a área de transferência",
 	"Copied shared chat URL to clipboard!": "URL do chat compartilhado copiada para a área de transferência!",
@@ -636,16 +636,16 @@
 	"Copy Link": "Copiar Link",
 	"Copy Path": "Copiar caminho",
 	"Copy Prompt": "Copiar prompt",
-	"Copy rich formatted content instead of plain text.": "",
+	"Copy rich formatted content instead of plain text.": "Copie conteúdo com formatação avançada em vez de texto simples.",
 	"Copy Share Link": "Copiar link de compartilhamento",
-	"Copy the current session token for authenticated requests.": "",
-	"Copy the latest assistant response when it completes.": "",
+	"Copy the current session token for authenticated requests.": "Copie o token da sessão atual para solicitações autenticadas.",
+	"Copy the latest assistant response when it completes.": "Copie a resposta mais recente do assistente quando ela for concluída.",
 	"Copy to clipboard": "Copiar para a área de transferência",
 	"Copy Token": "Copiar Token",
 	"Copy URL": "Copiar URL",
 	"Copying to clipboard was successful!": "Cópia para a área de transferência bem-sucedida!",
 	"CORS must be properly configured by the provider to allow requests from Open WebUI.": "O CORS deve ser configurado corretamente pelo provedor para permitir solicitações do Open WebUI.",
-	"Could not load release notes.": "",
+	"Could not load release notes.": "Não foi possível carregar as notas da versão.",
 	"Could not read file.": "Não foi possível ler o arquivo.",
 	"CPU": "CPU",
 	"Create": "Criar",
@@ -655,53 +655,53 @@
 	"Create Account": "Criar Conta",
 	"Create Admin Account": "Criar Conta de Administrador",
 	"Create and manage scheduled automations": "Criar e gerenciar automações agendadas",
-	"Create announcements shown to users in the app.": "",
+	"Create announcements shown to users in the app.": "Crie anúncios exibidos aos usuários no aplicativo.",
 	"Create Channel": "Criar Canal",
-	"Create chat tags from conversation content.": "",
+	"Create chat tags from conversation content.": "Crie tags de chat a partir do conteúdo da conversa.",
 	"Create Folder": "Criar Pasta",
 	"Create Image": "Criar imagem",
 	"Create Knowledge": "Criar Base de Conhecimento",
-	"Create missing groups from OAuth claims.": "",
+	"Create missing groups from OAuth claims.": "Crie os grupos ausentes a partir das claims do OAuth.",
 	"Create Model": "Criar modelo",
 	"Create new key": "Criar nova chave",
 	"Create new secret key": "Criar nova chave secreta",
-	"Create Prompt": "",
+	"Create Prompt": "Criar prompt",
 	"Create scheduled prompts that run automatically on a recurring basis.": "Crie prompts agendados que sejam executados automaticamente de forma recorrente.",
-	"Create your first note from the Create menu.": "",
-	"Create, copy, or rotate your API key.": "",
-	"Created": "",
+	"Create your first note from the Create menu.": "Crie sua primeira nota no menu Criar.",
+	"Create, copy, or rotate your API key.": "Crie, copie ou altere sua chave de API.",
+	"Created": "Criado",
 	"Created at": "Criado em",
 	"Created At": "Criado Em",
 	"Created by": "Criado por",
 	"Created by you": "Criado por você",
 	"Created on {{date}}": "Criado em {{date}}",
-	"Credential used for Document Intelligence.": "",
-	"Credentials used to authenticate with the Jupyter server.": "",
+	"Credential used for Document Intelligence.": "Credencial usada no Document Intelligence.",
+	"Credentials used to authenticate with the Jupyter server.": "Credenciais usadas para autenticação com o servidor Jupyter.",
 	"CSV Import": "Importação CSV",
 	"Ctrl+Enter to Send": "Ctrl+Enter para enviar",
-	"Current": "",
+	"Current": "Atual",
 	"Current Model": "Modelo Atual",
 	"Current Password": "Senha Atual",
-	"Current streak": "",
+	"Current streak": "Sequência atual",
 	"Custom": "Personalizado",
 	"Custom color": "Cor personalizada",
 	"Custom description enabled": "Descrição personalizada habilitada",
 	"Custom Gender": "Gênero personalizado",
-	"Custom message shown while an account waits for approval.": "",
+	"Custom message shown while an account waits for approval.": "Mensagem personalizada exibida enquanto uma conta aguarda aprovação.",
 	"Custom Parameter Name": "Nome do parâmetro personalizado",
 	"Custom Parameter Value": "Valor do parâmetro personalizado",
-	"Custom permissions": "",
+	"Custom permissions": "Permissões personalizadas",
 	"Custom range": "Intervalo personalizado",
-	"Custom title shown while an account waits for approval.": "",
+	"Custom title shown while an account waits for approval.": "Título personalizado exibido enquanto uma conta aguarda aprovação.",
 	"Daily": "Diário",
-	"Daily at": "",
+	"Daily at": "Diariamente às",
 	"Daily Messages": "Mensagens Diárias",
 	"Danger Zone": "Zona de Perigo",
 	"Dark": "Escuro",
 	"Data Controls": "Controle de Dados",
 	"Database": "Banco de Dados",
 	"Datalab Marker API": "API do Marcador do Datalab",
-	"Datalab Marker service endpoint used for document parsing.": "",
+	"Datalab Marker service endpoint used for document parsing.": "Endpoint do serviço Datalab Marker usado para análise de documentos.",
 	"Date Modified": "Data de Modificação",
 	"Day": "Dia",
 	"DD/MM/YYYY": "DD/MM/AAAA",
@@ -728,8 +728,8 @@
 	"Default User Role": "Padrão para novos usuários",
 	"Default webhook": "Webhook padrão",
 	"Defaults": "Padrões",
-	"Defines the behavior used for spoken conversations.": "",
-	"Delegate focused work to parallel sub-agents": "",
+	"Defines the behavior used for spoken conversations.": "Define o comportamento usado em conversas faladas.",
+	"Delegate focused work to parallel sub-agents": "Delegue trabalhos específicos a subagentes paralelos",
 	"Delete": "Excluir",
 	"Delete {{name}}": "Excluir {{name}}",
 	"Delete a model": "Excluir um modelo",
@@ -756,7 +756,7 @@
 	"Delete skill?": "Excluir skill?",
 	"delete this link": "Excluir este link",
 	"Delete tool?": "Excluir ferramenta?",
-	"Delete uploaded files from the upload directory.": "",
+	"Delete uploaded files from the upload directory.": "Exclua os arquivos enviados do diretório de uploads.",
 	"Delete User": "Excluir Usuário",
 	"Delete Version": "Excluir versão",
 	"Deleted": "Excluído",
@@ -773,7 +773,7 @@
 	"Description": "Descrição",
 	"Deselect": "Desmarcar",
 	"Detect Artifacts Automatically": "Detectar artefatos automaticamente",
-	"Detect generated artifacts and show them in the artifact workspace.": "",
+	"Detect generated artifacts and show them in the artifact workspace.": "Detecte os artefatos gerados e exiba-os no workspace de artefatos.",
 	"Dictate": "Ditar",
 	"Didn't fully follow instructions": "Não seguiu completamente as instruções",
 	"Direct": "Direto",
@@ -787,7 +787,7 @@
 	"Directory name": "Nome do diretório",
 	"Directory renamed.": "Diretório renomeado.",
 	"Directory selection was cancelled": "A seleção do diretório foi cancelada",
-	"Disable all": "",
+	"Disable all": "Desativar tudo",
 	"Disable All": "Desativar tudo",
 	"Disable Code Interpreter": "Desativar o interpretador de código",
 	"Disable Image Extraction": "Desativar extração de imagem",
@@ -809,47 +809,47 @@
 	"Display Chat Title in Tab": "Exibir título do chat na aba",
 	"Display Emoji in Call": "Exibir Emoji na Chamada",
 	"Display Multi-model Responses in Tabs": "Exibir respostas de vários modelos em guias",
-	"Display name for this LDAP connection.": "",
-	"Display name shown for the OAuth provider.": "",
+	"Display name for this LDAP connection.": "Nome de exibição desta conexão LDAP.",
+	"Display name shown for the OAuth provider.": "Nome de exibição mostrado para o provedor OAuth.",
 	"Display the Username Instead of You in the Chat": "Exibir o nome de usuário em vez de Você no Chat",
 	"Displays citations in the response": "Exibir citações na resposta",
 	"Displays status updates (e.g., web search progress) in the response": "Exibe atualizações de status (por exemplo, progresso da pesquisa na web) na resposta",
-	"Do not extract images from PDFs during Marker processing.": "",
+	"Do not extract images from PDFs during Marker processing.": "Não extraia imagens de PDFs durante o processamento pelo Marker.",
 	"Do not install functions from sources you do not fully trust.": "Não instale funções de origens nas quais você não confia totalmente.",
 	"Do not install tools from sources you do not fully trust.": "Não instale ferramentas de origens nas quais você não confia totalmente.",
 	"Do you want to sync your usage stats with Open WebUI Community?": "Deseja sincronizar suas estatísticas de uso com a Comunidade Open WebUI?",
 	"Docling": "Docling",
 	"Docling Parameters": "Parâmetros Docling",
-	"Docling Server URL": "",
+	"Docling Server URL": "URL do servidor Docling",
 	"Docling Server URL required.": "URL do servidor Docling necessária.",
-	"Docling service endpoint used for parsing.": "",
+	"Docling service endpoint used for parsing.": "Endpoint do serviço Docling usado para análise.",
 	"Document": "Documento",
 	"Document ID Field": "Campo de ID do Documento",
 	"Document Intelligence": "Inteligência de documentos",
 	"Document Intelligence endpoint required.": "É necessário o endpoint do Document Intelligence.",
-	"Document Intelligence endpoint used for parsing.": "",
+	"Document Intelligence endpoint used for parsing.": "Endpoint do Document Intelligence usado para análise.",
 	"Document Intelligence Model": "Modelo de Inteligência de Documentos",
-	"Document Loader URL": "",
+	"Document Loader URL": "URL do carregador de documentos",
 	"Documentation": "Documentação",
 	"Documents": "Documentos",
 	"does not make any external connections, and your data stays securely on your locally hosted server.": "não faz nenhuma conexão externa, e seus dados permanecem seguros no seu servidor local.",
 	"Domain Filter List": "Lista de Filtros de Domínio",
-	"Don't fetch random pipelines from sources you don't trust.": "",
+	"Don't fetch random pipelines from sources you don't trust.": "Não obtenha pipelines aleatórios de fontes nas quais você não confia.",
 	"Don't have an account?": "Não tem uma conta?",
 	"Don't like the style": "Não gostei do estilo",
 	"Done": "Concluído",
 	"Download": "Baixar",
 	"Download & Delete": "Baixar e excluir",
-	"Download all users as CSV.": "",
+	"Download all users as CSV.": "Baixe todos os usuários como CSV.",
 	"Download as JSON": "Baixar como JSON",
 	"Download as SVG": "Baixar como SVG",
 	"Download canceled": "Download cancelado",
-	"Download every user's chat history as JSON.": "",
-	"Download the application database when supported.": "",
-	"Download the current admin configuration as JSON.": "",
-	"Download your chat history as a JSON export.": "",
+	"Download every user's chat history as JSON.": "Baixe o histórico de chat de todos os usuários como JSON.",
+	"Download the application database when supported.": "Baixe o banco de dados do aplicativo quando houver suporte.",
+	"Download the current admin configuration as JSON.": "Baixe a configuração atual do administrador como JSON.",
+	"Download your chat history as a JSON export.": "Baixe seu histórico de chat como uma exportação JSON.",
 	"Downloading stats...": "Baixando estatísticas...",
-	"Drag to reorder": "",
+	"Drag to reorder": "Arraste para reordenar",
 	"Draw": "Empate",
 	"Drop any files here to upload": "Solte qualquer arquivo aqui para fazer upload",
 	"Drop files here": "Solte os arquivos aqui",
@@ -900,8 +900,8 @@
 	"ElevenLabs": "ElevenLabs",
 	"Email": "Email",
 	"Email Claim": "Reivindicação de Email",
-	"Email domains allowed to sign in with OAuth.": "",
-	"Email shown in the pending account overlay.": "",
+	"Email domains allowed to sign in with OAuth.": "Domínios de e-mail autorizados a entrar com OAuth.",
+	"Email shown in the pending account overlay.": "E-mail exibido na sobreposição de conta pendente.",
 	"Embedding": "Embedding",
 	"Embedding Batch Size": "Tamanho do Lote de Embedding",
 	"Embedding Concurrent Requests": "Solicitações Simultâneas de Embedding",
@@ -912,7 +912,7 @@
 	"Empty": "Vazio",
 	"Empty message": "Mensagem vazia",
 	"Enable All": "Ativar tudo",
-	"Enable background sub-agents": "",
+	"Enable background sub-agents": "Habilitar subagentes em segundo plano",
 	"Enable Code Execution": "Habilitar execução de código",
 	"Enable Code Interpreter": "Habilitar intérprete de código",
 	"Enable Image Generation": "Habilitar Geração de Imagem",
@@ -920,7 +920,7 @@
 	"Enable Memory Mapping (mmap) to load model data. This option allows the system to use disk storage as an extension of RAM by treating disk files as if they were in RAM. This can improve model performance by allowing for faster data access. However, it may not work correctly with all systems and can consume a significant amount of disk space.": "Habilite o mapeamento de memória (mmap) para carregar dados do modelo. Esta opção permite que o sistema use o armazenamento em disco como uma extensão da RAM, tratando os arquivos do disco como se estivessem na RAM. Isso pode melhorar o desempenho do modelo, permitindo acesso mais rápido aos dados. No entanto, pode não funcionar corretamente com todos os sistemas e consumir uma quantidade significativa de espaço em disco.",
 	"Enable Message Queue": "Habilitar fila de mensagens",
 	"Enable Mirostat sampling for controlling perplexity.": "Habilitar amostragem Mirostat para controlar a perplexidade.",
-	"Enable sub-agents": "",
+	"Enable sub-agents": "Habilitar subagentes",
 	"Enable Web Search": "Habilitar Pesquisa na Web",
 	"Enable, disable, or customize the reasoning tags used by the model. \"Enabled\" uses default tags, \"Disabled\" turns off reasoning tags, and \"Custom\" lets you specify your own start and end tags.": "Habilite, desabilite ou personalize as tags de raciocínio usadas pelo modelo. \"Ativado\" usa tags padrão, \"Desativado\" desativa as tags de raciocínio e \"Personalizado\" permite que você especifique suas próprias tags de início e fim.",
 	"Enabled": "Ativado",
@@ -929,7 +929,7 @@
 	"Endpoint URL": "URL do Endpoint",
 	"Enforce Temporary Chat": "Aplicar chat temporário",
 	"Enhance": "Melhorar",
-	"Enhance this note and update it.": "",
+	"Enhance this note and update it.": "Aprimore esta nota e atualize-a.",
 	"Enrich Hybrid Search Text": "Enriquecer o texto da pesquisa híbrida",
 	"Ensure your CSV file includes 4 columns in this order: Name, Email, Password, Role.": "Certifique-se de que seu arquivo CSV inclua 4 colunas nesta ordem: Nome, Email, Senha, Função.",
 	"Enter {{role}} message here": "Digite a mensagem de {{role}} aqui",
@@ -937,7 +937,7 @@
 	"Enter a watermark for the response. Leave empty for none.": "Insira uma marca d'água para a resposta. Deixe em branco se não houver nenhuma.",
 	"Enter additional headers in JSON format": "Insira cabeçalhos adicionais no formato JSON",
 	"Enter additional headers in JSON format (e.g. {\"X-Custom-Header\": \"value\"}": "Insira cabeçalhos adicionais no formato JSON (por exemplo, {{'{{\"X-Custom-Header\": \"value\"}}'}})",
-	"Enter additional OpenAI-compatible TTS request parameters as JSON.": "",
+	"Enter additional OpenAI-compatible TTS request parameters as JSON.": "Insira parâmetros adicionais de solicitação TTS compatíveis com OpenAI em formato JSON.",
 	"Enter additional parameters in JSON format": "Insira parâmetros adicionais no formato JSON",
 	"Enter api auth string (e.g. username:password)": "Digite a string de autenticação da API (por exemplo, username:password)",
 	"Enter Application DN": "Digite o DN da Aplicação",
@@ -976,7 +976,7 @@
 	"Enter Firecrawl Timeout": "Insira o Tempo Limite do Firecrawl",
 	"Enter folder name": "Digite o nome da pasta",
 	"Enter function name filter list (e.g. func1, !func2)": "Insira a lista de filtros de nomes de funções (por exemplo, func1, !func2)",
-	"Enter GitHub Raw URL": "",
+	"Enter GitHub Raw URL": "Insira a URL Raw do GitHub",
 	"Enter Google PSE API Key": "Digite a Chave API do Google PSE",
 	"Enter Google PSE Engine Id": "Digite o ID do Motor do Google PSE",
 	"Enter hex color (e.g. #FF0000)": "Digite a cor hexadecimal (por exemplo, #FF0000)",
@@ -1067,8 +1067,8 @@
 	"Enter Your Password": "Digite Sua Senha",
 	"Enter Your Role": "Digite Sua Função",
 	"Enter Your Username": "Digite seu usuário",
-	"Enterprise licenses also include dedicated support, customization options, and more, at a fraction of the cost of building and maintaining this stack internally.": "",
-	"Enterprise licensing": "",
+	"Enterprise licenses also include dedicated support, customization options, and more, at a fraction of the cost of building and maintaining this stack internally.": "As licenças empresariais também incluem suporte dedicado, opções de personalização e muito mais, por uma fração do custo de criar e manter esta stack internamente.",
+	"Enterprise licensing": "Licenciamento empresarial",
 	"Entra ID": "ID Entra",
 	"Environment Variables": "Variáveis de Ambiente",
 	"Ephemeral": "Efêmero",
@@ -1091,13 +1091,13 @@
 	"Event title": "Título do evento",
 	"Event updated": "Evento atualizado",
 	"Events": "Eventos",
-	"Every {{count}} hours_one": "",
-	"Every {{count}} hours_many": "",
-	"Every {{count}} hours_other": "",
-	"Every {{count}} minutes_one": "",
-	"Every {{count}} minutes_many": "",
-	"Every {{count}} minutes_other": "",
-	"Every minute": "",
+	"Every {{count}} hours_one": "A cada {{count}} hora",
+	"Every {{count}} hours_many": "A cada {{count}} horas",
+	"Every {{count}} hours_other": "A cada {{count}} horas",
+	"Every {{count}} minutes_one": "A cada {{count}} minuto",
+	"Every {{count}} minutes_many": "A cada {{count}} minutos",
+	"Every {{count}} minutes_other": "A cada {{count}} minutos",
+	"Every minute": "A cada minuto",
 	"Exa API Key": "Chave da API Exa",
 	"Example": "Exemplo",
 	"Example: (&(objectClass=inetOrgPerson)(uid=%s))": "Exemplo: (&(objectClass=inetOrgPerson)(uid=%s))",
@@ -1109,7 +1109,7 @@
 	"Exclude": "Excluir",
 	"Execute code": "Executar código",
 	"Execute code for analysis": "Executar código para análise",
-	"Executing {{NAME}}...": "",
+	"Executing {{NAME}}...": "Executando {{NAME}}...",
 	"Expand": "Expandir",
 	"Experimental": "Experimental",
 	"Explain": "Explicar",
@@ -1122,7 +1122,7 @@
 	"Export chat (.json)": "Exportar chat (.json)",
 	"Export Chats": "Exportar Chats",
 	"Export Config": "Exportar Configuração",
-	"Export JSON": "",
+	"Export JSON": "Exportar JSON",
 	"Export Models": "Exportar Modelos",
 	"Export Prompts": "Exportar Prompts",
 	"Export Skills": "Exportar Skills",
@@ -1133,8 +1133,8 @@
 	"External Document Loader URL required.": "URL do carregador de documentos externo necessária.",
 	"External Knowledge Source": "Fonte de Conhecimento Externa",
 	"External Knowledge Sources": "Fontes de Conhecimento Externas",
-	"External reranker endpoint.": "",
-	"External service endpoint used to load document content.": "",
+	"External reranker endpoint.": "Endpoint do reranqueador externo.",
+	"External service endpoint used to load document content.": "Endpoint do serviço externo usado para carregar o conteúdo de documentos.",
 	"External Task Model": "Modelo de Tarefa Externa",
 	"External Tool Servers": "Servidores de Ferramentas Externas",
 	"External vectors must be generated with the same embedding model configured in Open WebUI.": "Os vetores externos devem ser gerados com o mesmo modelo de embedding configurado no OWUI.",
@@ -1142,10 +1142,10 @@
 	"External Web Loader URL": "URL do carregador da Web externo",
 	"External Web Search API Key": "Chave de API de pesquisa na Web externa",
 	"External Web Search URL": "URL de pesquisa na Web externa",
-	"Extract action items from this note.": "",
-	"Extract images from PDFs so OCR can process image-only pages.": "",
+	"Extract action items from this note.": "Extraia os itens de ação desta nota.",
+	"Extract images from PDFs so OCR can process image-only pages.": "Extraia imagens de PDFs para que o OCR possa processar páginas que contêm apenas imagens.",
 	"Fade Effect for Streaming Text": "Efeito de Fade para texto em streaming",
-	"Fade streaming text as it arrives.": "",
+	"Fade streaming text as it arrives.": "Aplique um efeito de esmaecimento ao texto em streaming conforme ele chega.",
 	"Failed to add file.": "Falha ao adicionar arquivo.",
 	"Failed to add members": "Falha ao adicionar membros",
 	"Failed to archive chat.": "Falha ao arquivar o chat.",
@@ -1156,7 +1156,7 @@
 	"Failed to connect to {{URL}} terminal server": "Falha ao conectar ao servidor de terminal {{URL}}",
 	"Failed to copy link": "Falha ao copiar o link",
 	"Failed to create API Key.": "Falha ao criar a Chave API.",
-	"Failed to create chat": "",
+	"Failed to create chat": "Falha ao criar o chat",
 	"Failed to delete calendar": "Falha ao excluir calendário",
 	"Failed to delete note": "Falha ao excluir a nota",
 	"Failed to delete webhook": "Falha ao excluir webhook",
@@ -1165,7 +1165,7 @@
 	"Failed to extract content from the file: {{error}}": "Falha ao extrair conteúdo do arquivo: {{error}}",
 	"Failed to extract content from the file.": "Falha ao extrair conteúdo do arquivo.",
 	"Failed to fetch models": "Falha ao buscar modelos",
-	"Failed to fork chat": "",
+	"Failed to fork chat": "Falha ao bifurcar o chat",
 	"Failed to generate title": "Falha ao gerar título",
 	"Failed to import models": "Falha ao importar modelos",
 	"Failed to load chat": "Falha ao carregar chat",
@@ -1174,9 +1174,9 @@
 	"Failed to load Excel/CSV file. Please try downloading it instead.": "Não foi possível carregar o arquivo Excel/CSV. Tente baixá-lo em vez disso.",
 	"Failed to load file content.": "Falha ao carregar o conteúdo do arquivo.",
 	"Failed to load Interface settings": "Falha ao carregar configurações da Interface",
-	"Failed to load policy: {{error}}": "",
+	"Failed to load policy: {{error}}": "Falha ao carregar a política: {{error}}",
 	"Failed to load PPTX file. Please try downloading it instead.": "Não foi possível carregar o arquivo PPTX. Tente baixá-lo em vez disso.",
-	"Failed to load usage": "",
+	"Failed to load usage": "Falha ao carregar o uso",
 	"Failed to move chat": "Falha ao mover o chat",
 	"Failed to process URL: {{url}}": "Falha ao processar URL: {{url}}",
 	"Failed to read clipboard contents": "Falha ao ler o conteúdo da área de transferência",
@@ -1204,7 +1204,7 @@
 	"Feedback Details": "Detalhes do comentário",
 	"Feel free to add specific details": "Sinta-se à vontade para adicionar detalhes específicos",
 	"Female": "Feminino",
-	"Fetch page contents without proxy environment variables.": "",
+	"Fetch page contents without proxy environment variables.": "Obtenha o conteúdo das páginas sem usar variáveis de ambiente de proxy.",
 	"Fetch URL Content Length Limit": "Limite de Tamanho do Conteúdo de URL",
 	"File": "Arquivo",
 	"File added successfully.": "Arquivo adicionado com sucesso.",
@@ -1259,13 +1259,13 @@
 	"Followed instructions perfectly": "Seguiu as instruções perfeitamente",
 	"for placeholders": "para espaços reservados",
 	"Force OCR": "Forçar OCR",
-	"Forking chat...": "",
+	"Forking chat...": "Bifurcando chat...",
 	"Form": "Formulário",
 	"Format Lines": "Formatar linhas",
-	"Format lines to detect inline math and styles.": "",
-	"Format Markdown in previews and compact content surfaces.": "",
-	"Format Markdown syntax in assistant responses.": "",
-	"Format Markdown syntax in your own messages.": "",
+	"Format lines to detect inline math and styles.": "Formate as linhas para detectar expressões matemáticas inline e estilos.",
+	"Format Markdown in previews and compact content surfaces.": "Formate Markdown em prévias e áreas de conteúdo compacto.",
+	"Format Markdown syntax in assistant responses.": "Formate a sintaxe Markdown nas respostas do assistente.",
+	"Format Markdown syntax in your own messages.": "Formate a sintaxe Markdown em suas próprias mensagens.",
 	"Formatting may be inconsistent from source.": "A formatação pode ser inconsistente em relação à fonte.",
 	"Forward": "Encaminhar",
 	"Forwards system user OAuth access token to authenticate": "Encaminha o token de acesso OAuth do usuário do sistema para autenticação",
@@ -1280,14 +1280,14 @@
 	"Function imported successfully": "Função importada com sucesso",
 	"Function is now globally disabled": "A função está agora desativada globalmente",
 	"Function is now globally enabled": "A função está agora ativada globalmente",
-	"Function Menu": "",
+	"Function Menu": "Menu de funções",
 	"Function Name": "Nome da Função",
 	"Function Name Filter List": "Lista de filtros de nomes de funções",
 	"Function starter": "Inicializador de função",
 	"Function updated successfully": "Função atualizada com sucesso",
 	"Functions": "Funções",
 	"Functions allow arbitrary code execution.": "Funções permitem a execução arbitrária de código.",
-	"Functions can execute arbitrary code.": "",
+	"Functions can execute arbitrary code.": "As funções podem executar código arbitrário.",
 	"Functions imported successfully": "Funções importadas com sucesso",
 	"Gemini": "Gemini",
 	"Gemini API Key is required.": "A chave da API Gemini é necessária.",
@@ -1296,22 +1296,22 @@
 	"General": "Geral",
 	"Generate": "Gerar",
 	"Generate an image": "Gerar uma imagem",
-	"Generate an image prompt before sending the request.": "",
+	"Generate an image prompt before sending the request.": "Gere um prompt de imagem antes de enviar a solicitação.",
 	"Generate and edit images": "Gere e edite imagens",
-	"Generate chat titles automatically from conversation content.": "",
+	"Generate chat titles automatically from conversation content.": "Gere títulos de chat automaticamente a partir do conteúdo da conversa.",
 	"Generate Message Pair": "Gerar par de mensagens",
-	"Generate suggested follow-up prompts after responses.": "",
-	"Generate tags for chats automatically.": "",
+	"Generate suggested follow-up prompts after responses.": "Gere sugestões de prompts de acompanhamento após as respostas.",
+	"Generate tags for chats automatically.": "Gere tags para os chats automaticamente.",
 	"Generated Image": "Imagem gerada",
 	"Generated images will appear here": "As imagens geradas aparecerão aqui.",
 	"Generating search query": "Gerando consulta de pesquisa",
 	"Generating...": "Gerando...",
-	"Generation": "",
+	"Generation": "Geração",
 	"Get current time and perform date/time calculations": "Obtenha a hora atual e realize cálculos de data/hora.",
 	"Get information on {{name}} in the UI": "Obtenha informações sobre {{name}} na interface",
 	"Get started": "Iniciar",
 	"Get started with {{WEBUI_NAME}}": "Iniciar com {{WEBUI_NAME}}",
-	"GitHub URL": "",
+	"GitHub URL": "URL do GitHub",
 	"Global": "Global",
 	"Good Response": "Boa Resposta",
 	"Google": "Google",
@@ -1323,21 +1323,21 @@
 	"Grokipedia": "Grokipedia",
 	"group": "grupo",
 	"Group": "Grupo",
-	"Group assigned to new users by default.": "",
+	"Group assigned to new users by default.": "Grupo atribuído por padrão aos novos usuários.",
 	"Group Channel": "Canal do grupo",
 	"Group Claim": "Reivindicação de Grupo",
 	"Group created successfully": "Grupo criado com sucesso",
 	"Group deleted successfully": "Grupo excluído com sucesso",
 	"Group Description": "Descrição do Grupo",
-	"Group Mapping": "",
-	"Group multi-model responses into tabs.": "",
+	"Group Mapping": "Mapeamento de grupos",
+	"Group multi-model responses into tabs.": "Agrupe as respostas de vários modelos em abas.",
 	"Group Name": "Nome do Grupo",
 	"Group updated successfully": "Grupo atualizado com sucesso",
 	"groups": "grupos",
 	"Groups": "Grupos",
-	"Guides how the assistant formats tool and function calls.": "",
-	"Guides inline completions while users type a message.": "",
-	"Guides the suggestions shown after an assistant response.": "",
+	"Guides how the assistant formats tool and function calls.": "Orienta como o assistente formata chamadas de ferramentas e funções.",
+	"Guides inline completions while users type a message.": "Orienta as conclusões inline enquanto os usuários digitam uma mensagem.",
+	"Guides the suggestions shown after an assistant response.": "Orienta as sugestões exibidas após uma resposta do assistente.",
 	"H1": "Título",
 	"H2": "Subtítulo",
 	"H3": "Sub-subtítulo",
@@ -1346,7 +1346,7 @@
 	"Headers": "Cabeçalhos",
 	"Headers must be a valid JSON object": "Os cabeçalhos devem ser um objeto JSON válido",
 	"Height": "Altura",
-	"Height in pixels to compress images to. Leave empty for no compression.": "",
+	"Height in pixels to compress images to. Leave empty for no compression.": "Altura em pixels para a qual as imagens serão comprimidas. Deixe vazio para não comprimir.",
 	"Hello, {{name}}": "Olá, {{name}}",
 	"Help": "Ajuda",
 	"Help the community discover great models": "Ajude a comunidade a descobrir ótimos modelos.",
@@ -1395,45 +1395,45 @@
 	"Image Size": "Tamanho da imagem",
 	"Images": "Imagens",
 	"Import": "Importar",
-	"Import admin configuration from a JSON export file.": "",
-	"Import chat history from a JSON export file.": "",
+	"Import admin configuration from a JSON export file.": "Importe a configuração do administrador de um arquivo de exportação JSON.",
+	"Import chat history from a JSON export file.": "Importe o histórico de chat de um arquivo de exportação JSON.",
 	"Import Chats": "Importar Chats",
 	"Import Config": "Importar Configuração",
 	"Import From Link": "Importar do link",
-	"Import JSON": "",
+	"Import JSON": "Importar JSON",
 	"Import Models": "Importar Modelos",
 	"Import Prompts": "Importar Prompts",
 	"Import Skills": "Importar Skills",
 	"Import successful": "Importação bem-sucedida",
 	"Import Tools": "Importar Ferramentas",
-	"Import txt/md": "",
+	"Import txt/md": "Importar txt/md",
 	"Important Update": "Atualização importante",
-	"Imported automations successfully": "",
-	"Imported notes successfully": "",
+	"Imported automations successfully": "Automações importadas com sucesso",
+	"Imported notes successfully": "Notas importadas com sucesso",
 	"Inactive": "Inativo",
 	"Include": "Incluir",
-	"Include saved memories in the system context.": "",
-	"Include voices that are not provided by a local speech service.": "",
-	"Increase contrast for controls and input surfaces.": "",
+	"Include saved memories in the system context.": "Inclua as memórias salvas no contexto do sistema.",
+	"Include voices that are not provided by a local speech service.": "Inclua vozes que não são fornecidas por um serviço de fala local.",
+	"Increase contrast for controls and input surfaces.": "Aumente o contraste dos controles e das áreas de entrada.",
 	"Increase UI Scale": "Aumentar a escala da interface do usuário",
 	"Influences how quickly the algorithm responds to feedback from the generated text. A lower learning rate will result in slower adjustments, while a higher learning rate will make the algorithm more responsive.": "Influencia a rapidez com que o algoritmo responde ao feedback do texto gerado. Uma taxa de aprendizado menor resultará em ajustes mais lentos, enquanto uma taxa de aprendizado maior tornará o algoritmo mais responsivo.",
 	"Info": "Informação",
 	"Initials": "Iniciais",
-	"Inject entire documents as context for comprehensive processing.": "",
+	"Inject entire documents as context for comprehensive processing.": "Injete documentos inteiros como contexto para um processamento abrangente.",
 	"Inject file content into conversation context": "Inserir conteúdo de arquivo no contexto da conversa",
 	"Inject the entire content as context for comprehensive processing, this is recommended for complex queries.": "Injete todo o conteúdo como contexto para processamento abrangente; isso é recomendado para consultas complexas.",
-	"Inject the entire content as context for comprehensive processing.": "",
+	"Inject the entire content as context for comprehensive processing.": "Injete todo o conteúdo como contexto para um processamento abrangente.",
 	"Input": "Entrada",
 	"Input Key (e.g. text, unet_name, steps)": "Chave de entrada (por exemplo, texto, unet_name, etapas)",
 	"Input Variables": "Variáveis de entrada",
 	"Insert": "Inserir",
 	"Insert Follow-Up Prompt to Input": "Inserir prompt de acompanhamento para entrada",
-	"Insert into note": "",
+	"Insert into note": "Inserir na nota",
 	"Insert Prompt as Rich Text": "Inserir prompt como texto enriquecido",
-	"Insert selected follow-up prompts directly into the composer.": "",
+	"Insert selected follow-up prompts directly into the composer.": "Insira os prompts de acompanhamento selecionados diretamente no editor.",
 	"Insert Suggestion Prompt to Input": "Inserir prompt de sugestão para entrada",
-	"Install": "",
-	"Installing": "",
+	"Install": "Instalar",
+	"Installing": "Instalando",
 	"Instant Auto-Send After Voice Transcription": "Envio Automático Instantâneo Após Transcrição de Voz",
 	"Instructions": "Instruções",
 	"Integration": "Integração",
@@ -1442,7 +1442,7 @@
 	"Interface Settings Access": "Acesso às configurações da interface",
 	"Invalid file format.": "Formato de arquivo inválido.",
 	"Invalid JSON file": "Arquivo JSON inválido",
-	"Invalid JSON format": "",
+	"Invalid JSON format": "Formato JSON inválido",
 	"Invalid JSON format for ComfyUI Edit Workflow.": "Formato JSON inválido para o fluxo de trabalho de edição do ComfyUI.",
 	"Invalid JSON format for ComfyUI Workflow.": "Formato JSON inválido para o fluxo de trabalho do ComfyUI.",
 	"Invalid JSON format for Parameters": "Formato JSON inválido para parâmetros",
@@ -1456,26 +1456,26 @@
 	"Jina API Key": "Chave de API Jina",
 	"join our Discord for help.": "junte-se ao nosso Discord para obter ajuda.",
 	"JSON": "JSON",
-	"JSON Base64": "",
-	"JSON options passed to Marker for advanced parsing behavior.": "",
+	"JSON Base64": "JSON Base64",
+	"JSON options passed to Marker for advanced parsing behavior.": "Opções JSON enviadas ao Marker para configurar o comportamento avançado de análise.",
 	"JSON Preview": "Pré-visualização JSON",
 	"JSON Spec": "Especificação JSON",
 	"July": "Julho",
 	"June": "Junho",
 	"Jupyter Auth": "Autenticação Jupyter",
-	"Jupyter Password": "",
-	"Jupyter Token": "",
+	"Jupyter Password": "Senha do Jupyter",
+	"Jupyter Token": "Token do Jupyter",
 	"Jupyter URL": "URL do Jupyter",
 	"JWT Expiration": "Expiração do JWT",
 	"JWT Token": "Token JWT",
 	"Kagi Search API Key": "Chave da API de pesquisa Kagi",
-	"Keep current webhook URL": "",
+	"Keep current webhook URL": "Manter a URL atual do webhook",
 	"Keep Follow-Up Prompts in Chat": "Mantenha prompts de acompanhamento no chat",
-	"Keep generated follow-up prompts visible in the chat.": "",
+	"Keep generated follow-up prompts visible in the chat.": "Mantenha os prompts de acompanhamento gerados visíveis no chat.",
 	"Keep in Sidebar": "Manter na barra lateral",
 	"Key": "Chave",
 	"Key is required": "Chave é obrigatória",
-	"Keyboard": "",
+	"Keyboard": "Teclado",
 	"Knowledge": "Conhecimento",
 	"Knowledge Access": "Acesso ao Conhecimento",
 	"Knowledge Base": "Base de Conhecimento",
@@ -1503,13 +1503,13 @@
 	"Last Active": "Última Atividade",
 	"Last Modified": "Última Modificação",
 	"Last reply": "Última resposta",
-	"Last run": "",
+	"Last run": "Última execução",
 	"LDAP": "LDAP",
-	"LDAP attribute used as the user email address.": "",
-	"LDAP attribute used as the username.": "",
-	"LDAP filter used to match signing-in users.": "",
-	"LDAP server hostname or IP address.": "",
-	"LDAP server port.": "",
+	"LDAP attribute used as the user email address.": "Atributo LDAP usado como endereço de e-mail do usuário.",
+	"LDAP attribute used as the username.": "Atributo LDAP usado como nome de usuário.",
+	"LDAP filter used to match signing-in users.": "Filtro LDAP usado para identificar os usuários que estão entrando.",
+	"LDAP server hostname or IP address.": "Nome do host ou endereço IP do servidor LDAP.",
+	"LDAP server port.": "Porta do servidor LDAP.",
 	"Leaderboard": "Tabela de classificação",
 	"Learn More": "Saiba Mais",
 	"Learn more about Open Terminal": "Saiba mais sobre o Open Terminal",
@@ -1517,7 +1517,7 @@
 	"Learn more about Voxtral transcription.": "Saiba mais sobre a transcrição do Voxtral.",
 	"Leave a public review for {{modelName}}": "Deixe uma avaliação pública para {{modelName}}",
 	"Leave empty for no compression": "Deixe em branco para nenhuma compactação",
-	"Leave empty for the built-in default.": "",
+	"Leave empty for the built-in default.": "Deixe vazio para usar o padrão integrado.",
 	"Leave empty for unlimited": "Deixe vazio para ilimitado",
 	"Leave empty to include all models from \"{{url}}\" endpoint": "Deixe em branco para incluir todos os modelos do endpoint \"{{url}}\"",
 	"Leave empty to include all models from \"{{url}}/api/tags\" endpoint": "Deixe em branco para incluir todos os modelos do endpoint \"{{url}}/api/tags\"",
@@ -1527,22 +1527,22 @@
 	"Leave empty to use the default config, or enter a valid json (see https://yandex.cloud/en/docs/search-api/api-ref/WebSearch/search#yandex.cloud.searchapi.v2.WebSearchRequest)": "Deixe em branco para usar a configuração padrão ou insira um JSON válido (consulte https://yandex.cloud/en/docs/search-api/api-ref/WebSearch/search#yandex.cloud.searchapi.v2.WebSearchRequest)",
 	"Leave empty to use the default model (voxtral-mini-latest).": "Deixe em branco para usar o modelo padrão (voxtral-mini-latest).",
 	"Leave empty to use the default prompt, or enter a custom prompt": "Deixe vazio para usar o prompt padrão, ou insira um prompt personalizado",
-	"Leave empty to use the default prompt, or enter a custom prompt.": "",
+	"Leave empty to use the default prompt, or enter a custom prompt.": "Deixe vazio para usar o prompt padrão ou insira um prompt personalizado.",
 	"Leave model field empty to use the default model.": "Deixe o campo do modelo vazio para usar o modelo padrão.",
 	"Legacy": "Legado",
-	"Let speech interrupt the assistant during a voice call.": "",
-	"Let users rate assistant responses.": "",
+	"Let speech interrupt the assistant during a voice call.": "Permita que a fala interrompa o assistente durante uma chamada de voz.",
+	"Let users rate assistant responses.": "Permita que os usuários avaliem as respostas do assistente.",
 	"lexical": "lexical",
 	"License": "Licença",
 	"Lifecycle JSON": "JSON de ciclo de vida",
-	"Lifetime tokens": "",
+	"Lifetime tokens": "Tokens de todo o período",
 	"Lift List": "Lista de elevação",
 	"Light": "Claro",
-	"Limit API keys to configured endpoints.": "",
+	"Limit API keys to configured endpoints.": "Limite as chaves de API aos endpoints configurados.",
 	"Limit concurrent search queries. 0 = unlimited (default). Set to 1 for sequential execution (recommended for APIs with strict rate limits like Brave free tier).": "Limitar consultas de pesquisa simultâneas. 0 = ilimitado (padrão). Defina como 1 para execução sequencial (recomendado para APIs com limites de taxa rígidos, como o nível gratuito do Brave).",
-	"Limit how much draft text is sent for suggestion generation.": "",
-	"Limit parallel web loader requests.": "",
-	"Link OAuth sign-ins to existing accounts with the same email.": "",
+	"Limit how much draft text is sent for suggestion generation.": "Limite a quantidade do texto de rascunho enviada para gerar sugestões.",
+	"Limit parallel web loader requests.": "Limite as solicitações paralelas do carregador da web.",
+	"Link OAuth sign-ins to existing accounts with the same email.": "Vincule entradas via OAuth a contas existentes com o mesmo e-mail.",
 	"Linkup API Key": "Chave da API Linkup",
 	"List": "Lista",
 	"List calendars, search, create, update, and delete calendar events": "Listar calendários, pesquisar, criar, atualizar e excluir eventos do calendário",
@@ -1554,24 +1554,24 @@
 	"Loaded": "Carregado",
 	"Loader": "Carregador",
 	"Loading Kokoro.js...": "Carregando Kokoro.js...",
-	"Loading policy...": "",
-	"Loading release notes...": "",
+	"Loading policy...": "Carregando política...",
+	"Loading release notes...": "Carregando notas da versão...",
 	"Loading...": "Carregando...",
 	"local": "local",
 	"Local": "Local",
 	"Local Task Model": "Modelo de Tarefa Local",
 	"Location": "Localização",
 	"Location access not allowed": "Acesso ao local não permitido",
-	"Longest active chat": "",
-	"Longest streak": "",
+	"Longest active chat": "Chat ativo mais longo",
+	"Longest streak": "Maior sequência",
 	"Lost": "Perdeu",
 	"Low": "Baixo",
 	"LTR": "Esquerda para Direita",
 	"Made by Open WebUI Community": "Feito pela Comunidade OpenWebUI",
-	"Make Default": "",
+	"Make Default": "Definir como padrão",
 	"Make password visible in the user interface": "Tornar a senha visível na interface do usuário",
-	"Make Private": "",
-	"Make Public": "",
+	"Make Private": "Tornar privado",
+	"Make Public": "Tornar público",
 	"Male": "Masculino",
 	"Manage": "Gerenciar",
 	"Manage Connections": "Gerenciar Conexões",
@@ -1581,35 +1581,35 @@
 	"Manage Ollama": "Gerenciar Ollama",
 	"Manage Ollama API Connections": "Gerenciar Conexões Ollama API",
 	"Manage OpenAI API Connections": "Gerenciar Conexões OpenAI API",
-	"Map OAuth claims to Open WebUI groups.": "",
-	"Map OAuth claims to Open WebUI roles.": "",
-	"Map workflow node inputs used for image edits.": "",
-	"Map workflow node inputs used for image generation.": "",
+	"Map OAuth claims to Open WebUI groups.": "Mapeie claims do OAuth para grupos do Open WebUI.",
+	"Map OAuth claims to Open WebUI roles.": "Mapeie claims do OAuth para funções do Open WebUI.",
+	"Map workflow node inputs used for image edits.": "Mapeie as entradas dos nós do fluxo de trabalho usadas na edição de imagens.",
+	"Map workflow node inputs used for image generation.": "Mapeie as entradas dos nós do fluxo de trabalho usadas na geração de imagens.",
 	"Mapped Source": "Fonte Mapeada",
 	"March": "Março",
 	"Markdown": "Markdown",
 	"Markdown Header Text Splitter": "Separador de texto de cabeçalho Markdown",
-	"Max background": "",
-	"Max concurrent": "",
-	"Max iterations": "",
-	"Max output": "",
+	"Max background": "Máximo em segundo plano",
+	"Max concurrent": "Máximo simultâneo",
+	"Max iterations": "Máximo de iterações",
+	"Max output": "Saída máxima",
 	"Max Speakers": "Máximo de locutores",
 	"Max tokens to retrieve (1024-32768, default 8192)": "Máximo de tokens para recuperar (1024-32768, padrão 8192)",
 	"Max Upload Count": "Quantidade máxima de anexos",
 	"Max Upload Size": "Tamanho máximo do arquivo",
 	"Maximum characters to return from fetched URLs. Leave empty for no limit.": "Número máximo de caracteres a retornar de URLs buscadas. Deixe em branco para sem limite.",
-	"Maximum concurrent embedding requests. Set to 0 for unlimited.": "",
-	"Maximum file size in MB. Leave empty for unlimited.": "",
+	"Maximum concurrent embedding requests. Set to 0 for unlimited.": "Número máximo de solicitações simultâneas de embedding. Defina como 0 para não limitar.",
+	"Maximum file size in MB. Leave empty for unlimited.": "Tamanho máximo do arquivo em MB. Deixe vazio para não limitar.",
 	"Maximum number of files allowed per folder.": "Número máximo de arquivos permitidos por pasta.",
 	"Maximum number of files per folder is {{max}}.": "O número máximo de arquivos por pasta é {{max}}.",
-	"Maximum number of files that can be used at once in chat.": "",
-	"Maximum number of hybrid results sent to the reranker.": "",
-	"Maximum number of retrieved chunks returned to the model.": "",
+	"Maximum number of files that can be used at once in chat.": "Número máximo de arquivos que podem ser usados de uma só vez no chat.",
+	"Maximum number of hybrid results sent to the reranker.": "Número máximo de resultados híbridos enviados ao reranqueador.",
+	"Maximum number of retrieved chunks returned to the model.": "Número máximo de chunks recuperados e retornados ao modelo.",
 	"Maximum of 3 models can be downloaded simultaneously. Please try again later.": "Máximo de 3 modelos podem ser baixados simultaneamente. Por favor, tente novamente mais tarde.",
-	"Maximum runtime in seconds before execution is stopped.": "",
-	"Maximum size for each text chunk.": "",
-	"Maximum time in seconds to wait for MinerU API responses.": "",
-	"Maximum time to wait while loading web content.": "",
+	"Maximum runtime in seconds before execution is stopped.": "Tempo máximo de execução em segundos antes que ela seja interrompida.",
+	"Maximum size for each text chunk.": "Tamanho máximo de cada chunk de texto.",
+	"Maximum time in seconds to wait for MinerU API responses.": "Tempo máximo de espera, em segundos, pelas respostas da API do MinerU.",
+	"Maximum time to wait while loading web content.": "Tempo máximo de espera durante o carregamento do conteúdo da web.",
 	"May": "Maio",
 	"MBR": "MBR",
 	"MCP": "MCP",
@@ -1625,18 +1625,18 @@
 	"Memory added successfully": "Memória adicionada com sucesso",
 	"Memory cleared successfully": "Memória limpa com sucesso",
 	"Memory deleted successfully": "Memória excluída com sucesso",
-	"Memory System Context": "",
+	"Memory System Context": "Contexto de memória do sistema",
 	"Memory updated successfully": "Memória atualizada com sucesso",
 	"Merge Accounts by Email": "Mesclar Contas por Email",
-	"Merge chunks smaller than this threshold when possible. Set to 0 to disable merging.": "",
+	"Merge chunks smaller than this threshold when possible. Set to 0 to disable merging.": "Quando possível, combine chunks menores que este limite. Defina como 0 para desativar a combinação.",
 	"Merge Responses": "Mesclar respostas",
 	"Merged Response": "Resposta Mesclada",
 	"Message": "Mensagem",
 	"Message counts and response timestamps": "Contagem de mensagens e registros de data e hora de resposta",
 	"Message counts are based on assistant responses.": "A contagem de mensagens é baseada nas respostas do assistente.",
-	"Message Rating": "",
+	"Message Rating": "Avaliação de mensagens",
 	"Message rating should be enabled to use this feature": "A avaliação de mensagens deve estar habilitada para usar este recurso",
-	"Message shown before web search runs.": "",
+	"Message shown before web search runs.": "Mensagem exibida antes da execução da pesquisa na web.",
 	"Message text...": "Texto da mensagem...",
 	"messages": "mensagens",
 	"Messages": "Mensagens",
@@ -1649,11 +1649,11 @@
 	"Microsoft Web IQ API Key": "Chave da API Microsoft Web IQ",
 	"min": "min",
 	"MinerU": "MinerU",
-	"MinerU API endpoint for the selected mode.": "",
+	"MinerU API endpoint for the selected mode.": "Endpoint da API do MinerU para o modo selecionado.",
 	"MinerU API Key required for Cloud API mode.": "Chave de API MinerU necessária para o modo Cloud API.",
 	"Mistral OCR": "Mistral OCR",
 	"Mistral OCR API Key required.": "Chave de API do Mistral OCR necessária.",
-	"Mistral OCR service endpoint.": "",
+	"Mistral OCR service endpoint.": "Endpoint do serviço Mistral OCR.",
 	"MistralAI": "MistralAI",
 	"Mo_day_of_week": "Seg",
 	"Model": "Modelo",
@@ -1667,8 +1667,8 @@
 	"Model {{name}} is now visible": "O modelo {{name}} agora está visível",
 	"Model accepts file inputs": "O modelo aceita entradas de arquivo",
 	"Model accepts image inputs": "Modelo aceita entradas de imagens",
-	"Model added to pinned models": "",
-	"Model added to selected models": "",
+	"Model added to pinned models": "Modelo adicionado aos modelos fixados",
+	"Model added to selected models": "Modelo adicionado aos modelos selecionados",
 	"Model can access Open Terminal for command execution and file management": "O modelo pode acessar o Open Terminal para execução de comandos e gerenciamento de arquivos",
 	"Model can execute code and perform calculations": "O modelo pode executar código e realizar cálculos",
 	"Model can generate images based on text prompts": "O modelo pode gerar imagens com base em prompts de texto",
@@ -1680,29 +1680,29 @@
 	"Model ID": "ID do Modelo",
 	"Model ID is required.": "É necessário o ID do modelo.",
 	"Model IDs": "IDs do modelo",
-	"Model is now private": "",
-	"Model is now public": "",
+	"Model is now private": "O modelo agora é privado",
+	"Model is now public": "O modelo agora é público",
 	"Model Name": "Nome do Modelo",
 	"Model name already exists, please choose a different one": "O nome do modelo já existe, escolha um diferente",
 	"Model Name is required.": "O nome do modelo é obrigatório.",
-	"Model name used by Document Intelligence.": "",
+	"Model name used by Document Intelligence.": "Nome do modelo usado pelo Document Intelligence.",
 	"Model names and usage frequency": "Nomes dos modelos e frequência de utilização",
 	"Model not found": "Modelo não encontrado",
 	"Model not selected": "Modelo não selecionado",
-	"Model order saved successfully": "",
-	"Model parameters": "",
+	"Model order saved successfully": "Ordem dos modelos salva com sucesso",
+	"Model parameters": "Parâmetros do modelo",
 	"Model Parameters": "Parâmetros do Modelo",
 	"Model Params": "Parâmetros do Modelo",
 	"Model Permissions": "Permissões do Modelo",
-	"Model removed from pinned models": "",
-	"Model removed from selected models": "",
+	"Model removed from pinned models": "Modelo removido dos modelos fixados",
+	"Model removed from selected models": "Modelo removido dos modelos selecionados",
 	"Model responses or outputs": "Respostas ou resultados do modelo",
 	"Model unloaded successfully": "Modelo descarregado com sucesso",
 	"Model updated successfully": "Modelo atualizado com sucesso",
 	"Model Usage": "Utilização do modelo",
-	"Model used to generate embeddings. Reindex knowledge after changing this.": "",
-	"Model used to rerank retrieved results.": "",
-	"Model-specific context compaction thresholds cannot exceed this token limit.": "",
+	"Model used to generate embeddings. Reindex knowledge after changing this.": "Modelo usado para gerar embeddings. Reindexe o conhecimento após alterá-lo.",
+	"Model used to rerank retrieved results.": "Modelo usado para reranquear os resultados recuperados.",
+	"Model-specific context compaction thresholds cannot exceed this token limit.": "Os limites de compactação de contexto específicos do modelo não podem exceder este limite de tokens.",
 	"Model(s) do not support file upload": "Modelo(s) não suportam upload de arquivo",
 	"Modelfile Content": "Conteúdo do Arquivo do Modelo",
 	"Models": "Modelos",
@@ -1720,11 +1720,11 @@
 	"More Concise": "Mais conciso",
 	"More options": "Mais opções",
 	"More Options": "Mais opções",
-	"Most used tools": "",
+	"Most used tools": "Ferramentas mais usadas",
 	"Move": "Mover",
-	"Move every chat into the archive after confirmation.": "",
+	"Move every chat into the archive after confirmation.": "Mova todos os chats para o arquivo após a confirmação.",
 	"Moved {{name}}": "{{name}} movido",
-	"Multipart Upload": "",
+	"Multipart Upload": "Upload multipart",
 	"Mute": "Silenciar",
 	"Muted": "Silenciado",
 	"My Terminal": "Meu Terminal",
@@ -1740,7 +1740,7 @@
 	"New Button": "Novo Botão",
 	"New calendar": "Novo calendário",
 	"New Calendar": "Novo Calendário",
-	"New chat": "",
+	"New chat": "Novo chat",
 	"New Chat": "Novo Chat",
 	"New directory": "Novo diretório",
 	"New Directory": "Novo Diretório",
@@ -1750,7 +1750,7 @@
 	"New Group": "Novo Grupo",
 	"New Note": "Nova nota",
 	"New Password": "Nova Senha",
-	"New Sign Ups": "",
+	"New Sign Ups": "Novos cadastros",
 	"New Temporary Chat": "Novo chat temporário",
 	"New Terminal": "Novo Terminal",
 	"New Tool": "Nova Ferramenta",
@@ -1759,15 +1759,15 @@
 	"Next message": "Próxima mensagem",
 	"Next run": "Próxima execução",
 	"No access grants. Private to you.": "Sem permissões de acesso. Privacidade exclusiva para você.",
-	"No actions found": "",
+	"No actions found": "Nenhuma ação encontrada",
 	"No activity data": "Sem dados de atividade",
 	"No additional headers are sent unless configured.": "Nenhum cabeçalho adicional é enviado a menos que configurado.",
 	"No authentication": "Sem autenticação",
 	"No automations found": "Nenhuma automação encontrada",
-	"No chat alerts": "",
-	"No chat history": "",
-	"No chat to compact": "",
-	"No chat to fork": "",
+	"No chat alerts": "Nenhum alerta de chat",
+	"No chat history": "Nenhum histórico de chat",
+	"No chat to compact": "Nenhum chat para compactar",
+	"No chat to fork": "Nenhum chat para bifurcar",
 	"No chats found": "Nenhum chat encontrado",
 	"No chats found for this user.": "Nenhum chat encontrado para este usuário.",
 	"No chats found.": "Nenhum chat encontrado.",
@@ -1777,7 +1777,7 @@
 	"No conversation to save": "Nenhuma conversa para salvar",
 	"No data": "Sem dados",
 	"No data found": "Nenhum dado encontrado",
-	"No description": "",
+	"No description": "Sem descrição",
 	"No distance available": "Sem distância disponível",
 	"No event webhooks configured.": "Nenhum webhook de evento configurado.",
 	"No expiration can pose security risks.": "A ausência de expiração pode representar riscos de segurança.",
@@ -1787,7 +1787,7 @@
 	"No files found": "Nenhum arquivo encontrado",
 	"No files in this knowledge base.": "Não existem arquivos nesta base de conhecimento.",
 	"No files yet. Upload files or run Python code to create them.": "Nenhum arquivo ainda. Envie arquivos ou execute código Python para criá-los.",
-	"No filters found": "",
+	"No filters found": "Nenhum filtro encontrado",
 	"No functions found": "Nenhuma função encontrada",
 	"No groups found": "Nenhum grupo encontrado",
 	"No history available": "Não há histórico disponível.",
@@ -1798,17 +1798,17 @@
 	"No knowledge bases found.": "Nenhuma base de conhecimento encontrada.",
 	"No knowledge found": "Nenhum conhecimento encontrado",
 	"No limit": "Sem limite",
-	"No matches": "",
-	"No model data": "",
+	"No matches": "Nenhuma correspondência",
+	"No model data": "Nenhum dado de modelo",
 	"No model IDs": "Nenhum ID de modelo",
-	"No model usage found": "",
+	"No model usage found": "Nenhum uso de modelo encontrado",
 	"No models accessible": "Nenhum modelo acessível",
 	"No models available": "Nenhum modelo disponível",
 	"No models found": "Nenhum modelo encontrado",
 	"No models selected": "Nenhum modelo selecionado",
 	"No Notes": "Sem Notas",
 	"No notes found": "Notas não encontradas",
-	"No notification targets configured.": "",
+	"No notification targets configured.": "Nenhum destino de notificação configurado.",
 	"No one": "Ninguém",
 	"No output items": "Nenhum item de saída",
 	"No pinned messages": "Nenhuma mensagem fixada",
@@ -1816,7 +1816,7 @@
 	"No Repeat": "Sem Repetição",
 	"No results": "Nenhum resultado encontrado",
 	"No results found": "Nenhum resultado encontrado",
-	"No runs yet": "",
+	"No runs yet": "Nenhuma execução ainda",
 	"No search query generated": "Nenhuma consulta de pesquisa gerada",
 	"No servers detected": "Nenhum servidor detectado",
 	"No skills found": "Nenhuma skill encontrada",
@@ -1828,7 +1828,7 @@
 	"No tool server connections configured.": "Nenhuma conexão de servidor de ferramentas configurada.",
 	"No tools accessible": "Nenhuma ferramenta acessível",
 	"No tools found": "Nenhuma ferramenta encontrada",
-	"No usage data found": "",
+	"No usage data found": "Nenhum dado de uso encontrado",
 	"No users were found.": "Nenhum usuário foi encontrado.",
 	"No valves": "Sem configurações",
 	"No valves to update": "Nenhuma configuração para atualizar",
@@ -1842,30 +1842,30 @@
 	"Not scheduled": "Não agendado",
 	"Note": "Nota",
 	"Note deleted successfully": "Nota excluída com sucesso",
-	"Note Menu": "",
+	"Note Menu": "Menu de notas",
 	"Notes": "Notas",
 	"Notes Public Sharing": "Compartilhamento Público das Notas",
 	"Notes Sharing": "Compartilhamento de notas",
-	"Nothing to compact": "",
+	"Nothing to compact": "Nada para compactar",
 	"Notification Sound": "Som de notificação",
-	"Notification Targets": "",
+	"Notification Targets": "Destinos de notificação",
 	"Notifications": "Notificações",
 	"November": "Novembro",
-	"Number of items processed per embedding batch.": "",
-	"Number of results processed per reranking batch.": "",
+	"Number of items processed per embedding batch.": "Número de itens processados por lote de embedding.",
+	"Number of results processed per reranking batch.": "Número de resultados processados por lote de reranqueamento.",
 	"OAuth": "OAuth",
 	"OAuth / OIDC": "OAuth / OIDC",
 	"OAuth 2.1": "OAuth 2.1",
 	"OAuth 2.1 (Static)": "OAuth 2.1 (Estático)",
-	"OAuth client identifier from the provider.": "",
-	"OAuth client secret from the provider.": "",
+	"OAuth client identifier from the provider.": "Identificador do cliente OAuth fornecido pelo provedor.",
+	"OAuth client secret from the provider.": "Segredo do cliente OAuth fornecido pelo provedor.",
 	"OAuth ID": "OAuth ID",
 	"OAuth Resource Parameter": "Parâmetro de Recurso OAuth",
 	"OAuth Scopes": "Escopos OAuth",
-	"OAuth scopes requested during sign-in.": "",
+	"OAuth scopes requested during sign-in.": "Escopos do OAuth solicitados durante a entrada.",
 	"OAuth Server URL": "URL do Servidor OAuth",
 	"OAuth session disconnected": "Sessão OAuth desconectada",
-	"OAuth Signup": "",
+	"OAuth Signup": "Cadastro via OAuth",
 	"October": "Outubro",
 	"Off": "Desligado",
 	"Okay, Let's Go!": "Ok, Vamos Lá!",
@@ -1875,7 +1875,7 @@
 	"Ollama API": "API Ollama",
 	"Ollama API settings updated": "Configurações da API Ollama atualizadas",
 	"Ollama Cloud API Key": "Chave da API Ollama Cloud",
-	"Ollama endpoint used for embeddings.": "",
+	"Ollama endpoint used for embeddings.": "Endpoint do Ollama usado para embeddings.",
 	"Ollama Version": "Versão Ollama",
 	"Omit": "Omitir",
 	"On": "Ligado",
@@ -1889,20 +1889,20 @@
 	"Only alphanumeric characters and hyphens are allowed in the command string.": "Apenas caracteres alfanuméricos e hífens são permitidos na string de comando.",
 	"Only can be triggered when the chat input is in focus.": "Só pode ser acionado quando o campo de entrada do chat estiver em foco.",
 	"Only collections can be edited, create a new knowledge base to edit/add documents.": "Somente coleções podem ser editadas. Crie uma nova base de conhecimento para editar/adicionar documentos.",
-	"Only install functions from sources you trust.": "",
-	"Only install tools from sources you trust.": "",
+	"Only install functions from sources you trust.": "Instale funções somente de fontes confiáveis.",
+	"Only install tools from sources you trust.": "Instale ferramentas somente de fontes confiáveis.",
 	"Only invited users can access": "Somente usuários convidados podem acessar.",
-	"Only return documents with a score greater than or equal to this value.": "",
+	"Only return documents with a score greater than or equal to this value.": "Retorne somente documentos com pontuação maior ou igual a este valor.",
 	"Only select users and groups with permission can access": "Somente usuários e grupos selecionados com permissão podem acessar.",
 	"Only sync new/updated chats": "Sincronizar apenas conversas novas/atualizadas",
-	"Only when away": "",
+	"Only when away": "Somente quando ausente",
 	"Oops! Looks like the URL is invalid. Please double-check and try again.": "Ops! Parece que a URL é inválida. Por favor, verifique novamente e tente de novo.",
 	"Oops! There are files still uploading. Please wait for the upload to complete.": "Ops! Ainda há arquivos sendo enviados. Por favor, aguarde a conclusão do envio.",
 	"Oops! There was an error in the previous response.": "Ops! Houve um erro na resposta anterior.",
 	"Oops! You're using an unsupported method (frontend only). Please serve the WebUI from the backend.": "Ops! Você está usando um método não suportado (somente frontend). Por favor, sirva a WebUI a partir do backend.",
-	"Open automation": "",
-	"Open create menu": "",
-	"Open detail blocks by default.": "",
+	"Open automation": "Abrir automação",
+	"Open create menu": "Abrir menu Criar",
+	"Open detail blocks by default.": "Abra os blocos de detalhes por padrão.",
 	"Open file": "Abrir arquivo",
 	"Open in full screen": "Abrir em tela cheia",
 	"Open in new tab": "Abrir em nova aba",
@@ -1911,13 +1911,13 @@
 	"Open Modal To Manage Floating Quick Actions": "Abra o Modal para gerenciar ações rápidas flutuantes",
 	"Open Modal To Manage Image Compression": "Abrir o Modal para gerenciar a compressão de imagens",
 	"Open Model Selector": "Abrir Seletor de Modelos",
-	"Open note": "",
+	"Open note": "Abrir nota",
 	"Open Settings": "Abrir configurações",
 	"Open Sidebar": "Abrir barra lateral",
 	"Open Terminal": "Open Terminal",
-	"Open the changelog modal after sign-in when enabled.": "",
-	"Open the file browser after selecting a terminal.": "",
-	"Open the file manager for uploaded files.": "",
+	"Open the changelog modal after sign-in when enabled.": "Quando habilitado, abra o modal do changelog após a entrada.",
+	"Open the file browser after selecting a terminal.": "Abra o navegador de arquivos após selecionar um terminal.",
+	"Open the file manager for uploaded files.": "Abra o gerenciador de arquivos para os arquivos enviados.",
 	"Open User Profile Menu": "Abrir menu de perfil do usuário",
 	"Open WebUI can use tools provided by any OpenAPI server.": "O Open WebUI pode usar ferramentas fornecidas por qualquer servidor OpenAPI.",
 	"Open WebUI uses faster-whisper internally.": "Open WebUI usa faster-whisper internamente.",
@@ -1929,14 +1929,14 @@
 	"OpenAI API Key is required.": "Chave API OpenAI é necessária.",
 	"OpenAI API settings updated": "Configurações OpenAI atualizadas",
 	"OpenAI URL/Key required.": "URL/Chave OpenAI necessária.",
-	"OpenAI-compatible embeddings endpoint.": "",
+	"OpenAI-compatible embeddings endpoint.": "Endpoint de embeddings compatível com OpenAI.",
 	"OpenAPI": "OpenAPI",
 	"OpenAPI Spec": "Especificação OpenAPI",
 	"openapi.json URL or Path": "URL ou caminho do openapi.json",
-	"OpenID discovery URL for this provider.": "",
+	"OpenID discovery URL for this provider.": "URL de descoberta OpenID deste provedor.",
 	"optional": "opcional",
 	"Optional": "Opcional",
-	"Optional API key for Ollama requests.": "",
+	"Optional API key for Ollama requests.": "Chave de API opcional para solicitações ao Ollama.",
 	"or": "ou",
 	"Ordered List": "Lista ordenada",
 	"Other": "Outro",
@@ -1945,23 +1945,23 @@
 	"OUTPUT": "SAÍDA",
 	"Output format": "Formato de saída",
 	"Output Format": "Formato de Saída",
-	"Overlap preserved between neighboring chunks.": "",
-	"Override the Gemini image generation endpoint.": "",
+	"Overlap preserved between neighboring chunks.": "Sobreposição preservada entre chunks adjacentes.",
+	"Override the Gemini image generation endpoint.": "Substitua o endpoint de geração de imagens do Gemini.",
 	"Overview": "Visão Geral",
 	"PaddleOCR-vl": "PaddleOCR-vl",
 	"PaddleOCR-vl API URL required.": "URL da API PaddleOCR-vl é obrigatória.",
-	"PaddleOCR-vl service endpoint.": "",
+	"PaddleOCR-vl service endpoint.": "Endpoint do serviço PaddleOCR-vl.",
 	"page": "página",
 	"Page": "Página",
-	"Page mode creates one document per page. Single mode keeps pages together for chunking across boundaries.": "",
+	"Page mode creates one document per page. Single mode keeps pages together for chunking across boundaries.": "O modo de página cria um documento por página. O modo único mantém as páginas juntas para permitir a divisão em chunks entre seus limites.",
 	"Paginate": "Paginar",
 	"Parameters": "Parâmetros",
 	"Parent message not found": "Mensagem principal não encontrada",
 	"Participate in community leaderboards and evaluations! Syncing aggregated usage stats helps drive research and improvements to Open WebUI. Your privacy is paramount: no message content is ever shared.": "Participe dos rankings e avaliações da comunidade! A sincronização de estatísticas de uso agregadas ajuda a impulsionar pesquisas e melhorias no Open WebUI. Sua privacidade é fundamental: nenhum conteúdo de mensagem é compartilhado.",
 	"Password": "Senha",
-	"Password for the bind DN.": "",
+	"Password for the bind DN.": "Senha do DN de vinculação.",
 	"Passwords do not match.": "As senhas não coincidem.",
-	"Paste inserted prompts as rich text when possible.": "",
+	"Paste inserted prompts as rich text when possible.": "Cole os prompts inseridos como texto rico quando possível.",
 	"Paste Large Text as File": "Cole Textos Longos como Arquivo",
 	"Path": "Caminho",
 	"Path copied": "Caminho copiado",
@@ -1970,13 +1970,13 @@
 	"PDF Extract Images (OCR)": "Extrair Imagens do PDF (OCR)",
 	"PDF Loader Mode": "Modo de carregamento de PDF",
 	"pdf, docx, pptx, xlsx": "pdf, docx, pptx, xlsx",
-	"Peak tokens": "",
+	"Peak tokens": "Pico de tokens",
 	"pending": "pendente",
 	"Pending": "Pendente",
 	"Pending Accounts": "Contas Pendentes",
 	"Pending User Overlay Content": "Conteúdo de sobreposição de usuário pendente",
 	"Pending User Overlay Title": "Título de sobreposição de usuário pendente",
-	"Permanently delete every chat after confirmation.": "",
+	"Permanently delete every chat after confirmation.": "Exclua permanentemente todos os chats após a confirmação.",
 	"Permission denied when accessing media devices": "Permissão negada ao acessar dispositivos de mídia",
 	"Permission denied when accessing microphone": "Permissão negada ao acessar o microfone",
 	"Permission denied when accessing microphone: {{error}}": "Permissão negada ao acessar o microfone: {{error}}",
@@ -1987,7 +1987,7 @@
 	"Perplexity Search API URL": "URL da API de pesquisa Perplexity",
 	"Perplexity Search Context Usage": "Uso do contexto de pesquisa do Perplexity",
 	"Persistent": "Persistente",
-	"Personal": "",
+	"Personal": "Pessoal",
 	"Personalization": "Personalização",
 	"Picture Claim": "Reivindicação de Foto",
 	"Pin": "Fixar",
@@ -1995,17 +1995,17 @@
 	"Pinned": "Fixado",
 	"Pinned Messages": "Mensagens fixadas",
 	"Pipe": "Pipe",
-	"Pipeline": "",
+	"Pipeline": "Pipeline",
 	"Pipeline deleted successfully": "Pipeline excluído com sucesso",
 	"Pipeline downloaded successfully": "Pipeline baixado com sucesso",
-	"Pipeline URL": "",
+	"Pipeline URL": "URL da pipeline",
 	"Pipelines": "Pipelines",
-	"Pipelines are a plugin system with arbitrary code execution.": "",
+	"Pipelines are a plugin system with arbitrary code execution.": "Pipelines são um sistema de plugins com execução de código arbitrário.",
 	"Pipelines Not Detected": "Pipelines Não Detectados",
-	"Place selected suggestion text into the composer.": "",
+	"Place selected suggestion text into the composer.": "Insira o texto da sugestão selecionada no editor.",
 	"Plain text (.md)": "Texto simples (.md)",
 	"Plain text (.txt)": "Texto simples (.txt)",
-	"Play assistant responses aloud automatically.": "",
+	"Play assistant responses aloud automatically.": "Reproduza automaticamente as respostas do assistente em voz alta.",
 	"Playground": "Playground",
 	"Playwright Timeout (ms)": "Tempo limite do Playwright (ms)",
 	"Playwright WebSocket URL": "URL do WebSocket do Playwright",
@@ -2028,18 +2028,18 @@
 	"Please select a valid JSON file": "Selecione um arquivo JSON válido",
 	"Please select at least one user for Direct Message channel.": "Por favor, selecione pelo menos um usuário para o canal de Mensagens Diretas.",
 	"Please wait until all files are uploaded.": "Aguarde até que todos os arquivos sejam enviados.",
-	"Policy changes apply to newly provisioned terminals. Refresh matching terminals to apply them to existing terminals.": "",
+	"Policy changes apply to newly provisioned terminals. Refresh matching terminals to apply them to existing terminals.": "As alterações de política se aplicam aos terminais recém-provisionados. Atualize os terminais correspondentes para aplicá-las aos terminais existentes.",
 	"Policy ID": "ID da Política",
 	"Policy ID is required": "ID da Política é obrigatório",
-	"Policy is still loading": "",
+	"Policy is still loading": "A política ainda está sendo carregada",
 	"Port": "Porta",
 	"Ports": "Portas",
 	"Positive attitude": "Atitude positiva",
 	"Prefer not to say": "Prefiro não dizer",
-	"Preferred transcript language codes, separated by commas.": "",
+	"Preferred transcript language codes, separated by commas.": "Códigos dos idiomas preferidos para transcrição, separados por vírgulas.",
 	"Prefix ID": "Prefixo ID",
 	"Prefix ID is used to avoid conflicts with other connections by adding a prefix to the model IDs - leave empty to disable": "O ID de prefixo é utilizado para evitar conflitos com outras conexões, adicionando um prefixo aos IDs dos modelos - deixe em branco para desativar.",
-	"Press keys...": "",
+	"Press keys...": "Pressione as teclas...",
 	"Prevent File Creation": "Impedir a criação de arquivos",
 	"Preview": "Visualização",
 	"Preview Access": "Acesso de Visualização",
@@ -2054,26 +2054,26 @@
 	"Prompt Autocompletion": "Preenchimento automático de prompts",
 	"Prompt Content": "Conteúdo do Prompt",
 	"Prompt created successfully": "Prompt criado com sucesso",
-	"Prompt Menu": "",
+	"Prompt Menu": "Menu de prompts",
 	"Prompt Name": "Nome do prompt",
 	"Prompt Suggestions": "Sugestões de Prompt",
 	"Prompt Template": "Modelo de Prompt",
-	"Prompt template used when retrieved context is injected.": "",
+	"Prompt template used when retrieved context is injected.": "Modelo de prompt usado quando o contexto recuperado é injetado.",
 	"Prompt updated successfully": "Prompt atualizado com sucesso",
 	"Prompts": "Prompts",
 	"Prompts Access": "Acesso aos Prompts",
 	"Prompts Public Sharing": "Compartilhamento Público dos Prompts",
 	"Prompts Sharing": "Compartilhamento de Prompts",
-	"Provide the --api-auth username and password when required.": "",
+	"Provide the --api-auth username and password when required.": "Forneça o nome de usuário e a senha de --api-auth quando necessário.",
 	"Provider": "Provedor",
-	"Provider groups blocked from signing in.": "",
+	"Provider groups blocked from signing in.": "Grupos do provedor impedidos de entrar.",
 	"Provider Name": "Nome do Provedor",
-	"Provider roles allowed to sign in.": "",
-	"Provider roles that grant admin access.": "",
+	"Provider roles allowed to sign in.": "Funções do provedor autorizadas a entrar.",
+	"Provider roles that grant admin access.": "Funções do provedor que concedem acesso de administrador.",
 	"Provider URL": "URL do Provedor",
-	"Provider used to generate document embeddings.": "",
-	"Provider used to rerank hybrid search results.": "",
-	"Proxy URL used for Youtube loader requests.": "",
+	"Provider used to generate document embeddings.": "Provedor usado para gerar embeddings de documentos.",
+	"Provider used to rerank hybrid search results.": "Provedor usado para reranquear os resultados da pesquisa híbrida.",
+	"Proxy URL used for Youtube loader requests.": "URL do proxy usada nas solicitações do carregador do YouTube.",
 	"Public": "Público",
 	"Pull \"{{searchValue}}\" from Ollama.com": "Obter \"{{searchValue}}\" de Ollama.com",
 	"Pull a model from Ollama.com": "Obter um modelo de Ollama.com",
@@ -2081,7 +2081,7 @@
 	"Pyodide file browser": "Navegador de arquivos Pyodide",
 	"Query Generation Prompt": "Prompt de Geração de Consulta",
 	"Querying": "Consultando",
-	"Queue outgoing messages instead of interrupting active responses.": "",
+	"Queue outgoing messages instead of interrupting active responses.": "Coloque as mensagens de saída na fila em vez de interromper as respostas ativas.",
 	"Quick Actions": "Ações rápidas",
 	"RAG Template": "Modelo RAG",
 	"Ran {{COUNT}} analyses": "Executou {{COUNT}} análises",
@@ -2093,13 +2093,13 @@
 	"Read more →": "Leia mais →",
 	"Read only": "Somente leitura",
 	"Read Only": "Somente leitura",
-	"Read the docs": "",
+	"Read the docs": "Leia a documentação",
 	"Read-Only Access": "Acesso somente leitura",
 	"Reason": "Razão",
 	"Reasoning Effort": "Esforço de raciocínio",
 	"Reasoning Tags": "Tags de raciocínio",
 	"Reasoning text...": "Texto de raciocínio...",
-	"Rebuild vectors for existing knowledge files.": "",
+	"Rebuild vectors for existing knowledge files.": "Reconstrua os vetores dos arquivos de conhecimento existentes.",
 	"Receives matching events across the instance, including system/config events and events associated with any user.": "Recebe eventos correspondentes em toda a instância, incluindo eventos de sistema/configuração e eventos associados a qualquer usuário.",
 	"Receives matching events that are not associated with a user.": "Recebe eventos correspondentes que não estão associados a um usuário.",
 	"Receives matching user-associated events only when the actor, user subject, or user data matches these users or current group members. System/config events are not sent.": "Recebe eventos de usuário correspondentes apenas quando o ator, assunto do usuário ou dados do usuário corresponderem a esses usuários ou membros atuais do grupo. Eventos de sistema/configuração não são enviados.",
@@ -2116,9 +2116,9 @@
 	"Refresh requested: {{count}} terminal(s)_many": "Atualização solicitada: {{count}} terminais",
 	"Refresh requested: {{count}} terminal(s)_other": "Atualização solicitada: {{count}} terminais",
 	"Refresh Terminals": "Atualizar Terminais",
-	"Refresh the account email from OAuth on sign-in.": "",
-	"Refresh the account name from OAuth on sign-in.": "",
-	"Refresh the profile picture from OAuth on sign-in.": "",
+	"Refresh the account email from OAuth on sign-in.": "Atualize o e-mail da conta pelo OAuth ao entrar.",
+	"Refresh the account name from OAuth on sign-in.": "Atualize o nome da conta pelo OAuth ao entrar.",
+	"Refresh the profile picture from OAuth on sign-in.": "Atualize a imagem de perfil pelo OAuth ao entrar.",
 	"Refreshing...": "Atualizando...",
 	"Refused when it shouldn't have": "Recusado quando não deveria",
 	"Regenerate": "Gerar novamente",
@@ -2140,14 +2140,14 @@
 	"Remove": "Remover",
 	"Remove {{MODELID}} from list.": "Remover {{MODELID}} da lista.",
 	"Remove action": "Remover ação",
-	"Remove existing OCR text and rerun OCR when Force OCR is off.": "",
+	"Remove existing OCR text and rerun OCR when Force OCR is off.": "Remova o texto de OCR existente e execute o OCR novamente quando Forçar OCR estiver desativado.",
 	"Remove file": "Remover arquivo",
 	"Remove File": "Remover Arquivo",
 	"Remove from favorites": "Remover dos favoritos",
 	"Remove image": "Remover imagem",
-	"Remove Pinned Model": "",
-	"Remove prompt suggestion": "",
-	"Remove Selected Model": "",
+	"Remove Pinned Model": "Remover modelo fixado",
+	"Remove prompt suggestion": "Remover sugestão de prompt",
+	"Remove Selected Model": "Remover modelo selecionado",
 	"Removing {{count}} stale files..._one": "Removendo {{count}} arquivo obsoleto...",
 	"Removing {{count}} stale files..._many": "Removendo {{count}} arquivos obsoletos...",
 	"Removing {{count}} stale files..._other": "Removendo {{count}} arquivos obsoletos...",
@@ -2156,13 +2156,13 @@
 	"Render Markdown in Assistant Messages": "Renderizar Markdown nas Mensagens do Assistente",
 	"Render Markdown in Previews": "Renderizar Markdown nas Pré-visualizações",
 	"Render Markdown in User Messages": "Renderizar Markdown nas Mensagens do Usuário",
-	"Render messages in compact bubble containers.": "",
+	"Render messages in compact bubble containers.": "Renderize as mensagens em contêineres de balões compactos.",
 	"Repeat": "Repetir",
 	"Reply": "Responder",
 	"Reply in Thread": "Responder no tópico",
 	"Reply to thread...": "Responder ao tópico...",
 	"Replying to {{NAME}}": "Respondendo para {{NAME}}",
-	"Request Format": "",
+	"Request Format": "Formato da solicitação",
 	"Require users to confirm before using Web Search.": "Exigir que os usuários confirmem antes de usar a Pesquisa na Web.",
 	"required": "obrigatório",
 	"Reranking Batch Size": "Tamanho do lote de reclassificação",
@@ -2172,7 +2172,7 @@
 	"Reset": "Redefinir",
 	"Reset All Models": "Redefinir todos os modelos",
 	"Reset all permissions to their initial configuration values": "Redefinir todas as permissões para seus valores de configuração inicial",
-	"Reset Defaults": "",
+	"Reset Defaults": "Redefinir padrões",
 	"Reset group permissions to match the current default user permissions": "Redefinir permissões do grupo para corresponder às permissões padrão atuais do usuário",
 	"Reset knowledge base?": "Redefinir base de conhecimento?",
 	"Reset persisted files": "Redefinir arquivos persistidos",
@@ -2182,11 +2182,11 @@
 	"Reset view": "Redefinir visualização",
 	"Response": "Resposta",
 	"Response notifications cannot be activated as the website permissions have been denied. Please visit your browser settings to grant the necessary access.": "Notificações de resposta não podem ser ativadas pois as permissões do site foram negadas. Por favor, visite as configurações do seu navegador para conceder o acesso necessário.",
-	"Response Splitting": "",
+	"Response Splitting": "Divisão de respostas",
 	"Response Watermark": "Marca d'água de resposta",
 	"Responses": "Respostas",
 	"Restart": "Reiniciar",
-	"Restrict or exclude domains using a comma-separated list.": "",
+	"Restrict or exclude domains using a comma-separated list.": "Restrinja ou exclua domínios usando uma lista separada por vírgulas.",
 	"Result": "Resultado",
 	"RESULT": "Resultado",
 	"Retrieval": "Recuperação",
@@ -2196,30 +2196,30 @@
 	"Retrieved {{count}} sources_many": "{{count}} fontes recuperadas",
 	"Retrieved {{count}} sources_other": "{{count}} fontes recuperadas",
 	"Retrieved 1 source": "1 fonte recuperada",
-	"Retry": "",
-	"Review and manage chats you have shared.": "",
-	"Rewrite the selected text.": "",
-	"Rewrite user requests for knowledge retrieval.": "",
-	"Rewrite user requests into web-search queries.": "",
-	"Rewrites user intent into an image-generation prompt.": "",
+	"Retry": "Tentar novamente",
+	"Review and manage chats you have shared.": "Revise e gerencie os chats que você compartilhou.",
+	"Rewrite the selected text.": "Reescreva o texto selecionado.",
+	"Rewrite user requests for knowledge retrieval.": "Reescreva as solicitações dos usuários para recuperação de conhecimento.",
+	"Rewrite user requests into web-search queries.": "Reescreva as solicitações dos usuários como consultas de pesquisa na web.",
+	"Rewrites user intent into an image-generation prompt.": "Reescreve a intenção do usuário como um prompt de geração de imagem.",
 	"Rich Text Input for Chat": "Entrada de rich text para o chat",
 	"Role": "Função",
-	"Role assigned to new users when they create an account.": "",
-	"Role Mapping": "",
+	"Role assigned to new users when they create an account.": "Função atribuída aos novos usuários quando eles criam uma conta.",
+	"Role Mapping": "Mapeamento de funções",
 	"Roles Claim": "Reivindicação de Funções",
 	"RTL": "Direita para Esquerda",
 	"Run": "Executar",
-	"Run AI on your own terms. Connect any model, extend with code, and protect what matters without compromise. Your models, your data, your machine, wherever you open it.": "",
+	"Run AI on your own terms. Connect any model, extend with code, and protect what matters without compromise. Your models, your data, your machine, wherever you open it.": "Execute IA do seu jeito. Conecte qualquer modelo, amplie com código e proteja o que importa sem abrir mão de nada. Seus modelos, seus dados e sua máquina, onde quer que você os acesse.",
 	"Run All": "Executar Tudo",
-	"Run embedding tasks concurrently to speed up processing.": "",
+	"Run embedding tasks concurrently to speed up processing.": "Execute tarefas de embedding simultaneamente para acelerar o processamento.",
 	"Run now": "Executar agora",
 	"Run Now": "Executar Agora",
-	"Run OCR on all PDF pages, even pages with embedded text.": "",
+	"Run OCR on all PDF pages, even pages with embedded text.": "Execute OCR em todas as páginas do PDF, mesmo nas que contêm texto incorporado.",
 	"Running": "Executando",
-	"Running Open WebUI for a team?": "",
+	"Running Open WebUI for a team?": "Usa o Open WebUI com uma equipe?",
 	"Running...": "Executando...",
-	"runs": "",
-	"Runs": "",
+	"runs": "execuções",
+	"Runs": "Execuções",
 	"Sa_day_of_week": "Sáb",
 	"Save": "Salvar",
 	"Save & Create": "Salvar e Criar",
@@ -2227,14 +2227,14 @@
 	"Save As Copy": "Salvar Como Cópia",
 	"Save Chat": "Salvar Chat",
 	"Saved": "Armazenado",
-	"Saved Memories": "",
+	"Saved Memories": "Memórias salvas",
 	"Saving chat logs directly to your browser's storage is no longer supported. Please take a moment to download and delete your chat logs by clicking the button below. Don't worry, you can easily re-import your chat logs to the backend through": "Salvar registros de chat diretamente no armazenamento do seu navegador não é mais suportado. Por favor, reserve um momento para baixar e excluir seus registros de chat clicando no botão abaixo. Não se preocupe, você pode facilmente reimportar seus registros de chat para o backend através de",
-	"Saving...": "",
+	"Saving...": "Salvando...",
 	"Schedule": "Agendar",
 	"Scheduled time must be in the future": "O horário agendado deve ser no futuro.",
 	"Scopes": "Escopos",
 	"Scroll On Branch Change": "Rolar na mudança de ramo",
-	"Scroll to the active branch when switching response branches.": "",
+	"Scroll to the active branch when switching response branches.": "Role até a ramificação ativa ao alternar entre ramificações de resposta.",
 	"Scroll to Top": "Rolar para o topo",
 	"Search": "Pesquisar",
 	"Search a model": "Pesquisar um modelo",
@@ -2259,7 +2259,7 @@
 	"Search Groups": "Pesquisar Grupos",
 	"Search In Models": "Pesquisar em modelos",
 	"Search Knowledge": "Pesquisar Conhecimento",
-	"Search Limits": "",
+	"Search Limits": "Limites de pesquisa",
 	"Search Memories": "Pesquisar Memórias",
 	"Search Models": "Pesquisar Modelos",
 	"Search Notes": "Pesquisar Notas",
@@ -2284,13 +2284,13 @@
 	"Searching the web": "Pesquisando na Internet...",
 	"Searxng Query URL": "URL da Consulta Searxng",
 	"Searxng search language (all, en, es, de, fr, etc.)": "Idioma de pesquisa do Searxng (all, en, es, de, fr, etc.)",
-	"Secrets": "",
+	"Secrets": "Segredos",
 	"See readme.md for instructions": "Veja readme.md para instruções",
 	"See what's new": "Veja o que há de novo",
 	"Seed": "Seed",
 	"Select": "Selecionar",
 	"Select {{modelName}} model": "Selecionar modelo {{modelName}}",
-	"Select a .py file": "",
+	"Select a .py file": "Selecione um arquivo .py",
 	"Select a base model (e.g. llama3, gpt-4o)": "Selecione um modelo base (por exemplo, llama3, gpt-4o)",
 	"Select a conversation to preview": "Selecione uma conversa para visualizar",
 	"Select a engine": "Selecione um motor",
@@ -2307,50 +2307,50 @@
 	"Select a theme": "Selecione um tema",
 	"Select a tool": "Selecione uma ferramenta",
 	"Select a voice": "Selecione uma voz",
-	"Select Action": "",
+	"Select Action": "Selecionar ação",
 	"Select All": "Selecionar Tudo",
 	"Select an auth method": "Selecione um método de autenticação",
 	"Select an embedding model engine": "Selecione um mecanismo de modelo de embedding",
 	"Select an engine": "Selecione um motor",
-	"Select an installed pipeline to configure or remove.": "",
+	"Select an installed pipeline to configure or remove.": "Selecione uma pipeline instalada para configurar ou remover.",
 	"Select an Ollama instance": "Selecione uma instância do Ollama",
 	"Select an option": "Selecione uma opção",
 	"Select an output format": "Selecione um formato de saída",
 	"Select dtype": "Selecionar dtype",
 	"Select Engine": "Selecionar Motor",
-	"Select Filter": "",
-	"Select how audio is sent to the OpenAI-compatible endpoint.": "",
-	"Select how Open WebUI authenticates with the Jupyter server.": "",
+	"Select Filter": "Selecionar filtro",
+	"Select how audio is sent to the OpenAI-compatible endpoint.": "Selecione como o áudio é enviado ao endpoint compatível com OpenAI.",
+	"Select how Open WebUI authenticates with the Jupyter server.": "Selecione como o Open WebUI se autentica com o servidor Jupyter.",
 	"Select how to split message text for TTS requests": "Selecione como dividir o texto da mensagem para solicitações TTS",
 	"Select Knowledge": "Selecionar Conhecimento",
 	"Select Method": "Selecione o método",
 	"Select model": "Selecione o modelo",
 	"Select only one model to call": "Selecione apenas um modelo para chamar",
-	"Select Skill": "",
-	"Select the Gemini endpoint method to call.": "",
-	"Select the local model precision used by Kokoro.js.": "",
-	"Select the Pipelines server to manage.": "",
-	"Select Tool": "",
+	"Select Skill": "Selecionar habilidade",
+	"Select the Gemini endpoint method to call.": "Selecione o método do endpoint Gemini a ser chamado.",
+	"Select the local model precision used by Kokoro.js.": "Selecione a precisão do modelo local usada pelo Kokoro.js.",
+	"Select the Pipelines server to manage.": "Selecione o servidor de Pipelines a ser gerenciado.",
+	"Select Tool": "Selecionar ferramenta",
 	"Select view": "Selecionar visualização",
-	"Selected": "",
+	"Selected": "Selecionado",
 	"Selected model: {{modelName}}": "Modelo selecionado: {{modelName}}",
 	"Selected model(s) do not support image inputs": "Modelo(s) selecionado(s) não suportam entradas de imagem",
 	"semantic": "semântica",
 	"Send": "Enviar",
 	"Send a Message": "Enviar uma Mensagem",
 	"Send events for": "Enviar eventos para",
-	"Send extra JSON parameters with each AUTOMATIC1111 request.": "",
-	"Send extra JSON parameters with each image generation request.": "",
+	"Send extra JSON parameters with each AUTOMATIC1111 request.": "Envie parâmetros JSON adicionais em cada solicitação ao AUTOMATIC1111.",
+	"Send extra JSON parameters with each image generation request.": "Envie parâmetros JSON adicionais em cada solicitação de geração de imagem.",
 	"Send message": "Enviar mensagem",
-	"Send notifications to configured webhook targets": "",
+	"Send notifications to configured webhook targets": "Enviar notificações aos destinos de webhook configurados",
 	"Send now": "Enviar agora",
-	"Send PDFs as base64 data URLs instead of uploading first.": "",
+	"Send PDFs as base64 data URLs instead of uploading first.": "Envie PDFs como URLs de dados base64 em vez de fazer o upload primeiro.",
 	"Send product events as JSON to external services. Chat destinations receive readable messages.": "Enviar eventos de produto como JSON para serviços externos. Destinos de chat recebem mensagens legíveis.",
-	"Send Test": "",
-	"Send transcribed voice input immediately after speech recognition finishes.": "",
-	"Send users directly to the OAuth provider from the sign-in page.": "",
+	"Send Test": "Enviar teste",
+	"Send transcribed voice input immediately after speech recognition finishes.": "Envie a entrada de voz transcrita assim que o reconhecimento de fala terminar.",
+	"Send users directly to the OAuth provider from the sign-in page.": "Envie os usuários diretamente ao provedor OAuth a partir da página de entrada.",
 	"Sends `stream_options: { include_usage: true }` in the request.\nSupported providers will return token usage information in the response when set.": "Envia `stream_options: { include_usage: true }` na solicitação. Provedores compatíveis retornarão informações sobre o uso de tokens na resposta quando configurado.",
-	"Separate output by page with page markers.": "",
+	"Separate output by page with page markers.": "Separe a saída por página usando marcadores de página.",
 	"September": "Setembro",
 	"SerpApi API Key": "Chave da API SerpApi",
 	"SerpApi Engine": "Motor SerpApi",
@@ -2364,23 +2364,23 @@
 	"Service Account": "Conta de Serviço",
 	"Session": "Sessão",
 	"Session expired. Please sign in again.": "Sessão expirada. Por favor, faça login novamente.",
-	"Set a local zoom level for the app interface.": "",
-	"Set a model-specific context compaction token threshold. When set, this overrides the global threshold up to the global cap.": "",
-	"Set a speech recognition language or leave it blank to detect automatically.": "",
+	"Set a local zoom level for the app interface.": "Defina um nível de zoom local para a interface do aplicativo.",
+	"Set a model-specific context compaction token threshold. When set, this overrides the global threshold up to the global cap.": "Defina um limite de tokens para compactação de contexto específico do modelo. Quando definido, ele substitui o limite global até o teto global.",
+	"Set a speech recognition language or leave it blank to detect automatically.": "Defina um idioma para reconhecimento de fala ou deixe em branco para detectar automaticamente.",
 	"Set as default": "Definir como padrão",
-	"Set as Pinned Model": "",
+	"Set as Pinned Model": "Definir como modelo fixado",
 	"Set as Production": "Definir como Produção",
-	"Set as Selected Model": "",
+	"Set as Selected Model": "Definir como modelo selecionado",
 	"Set embedding model": "Definir modelo de embedding",
 	"Set embedding model (e.g. {{model}})": "Definir modelo de embedding (por exemplo, {{model}})",
 	"Set reranking model (e.g. {{model}})": "Definir modelo de reclassificação (por exemplo, {{model}})",
-	"Set the birth date saved with your profile.": "",
-	"Set the default system prompt for new chats.": "",
-	"Set the display name shown across your account.": "",
+	"Set the birth date saved with your profile.": "Defina a data de nascimento salva em seu perfil.",
+	"Set the default system prompt for new chats.": "Defina o prompt do sistema padrão para novos chats.",
+	"Set the display name shown across your account.": "Defina o nome de exibição mostrado em toda a sua conta.",
 	"Set the number of layers, which will be off-loaded to GPU. Increasing this value can significantly improve performance for models that are optimized for GPU acceleration but may also consume more power and GPU resources.": "Defina o número de camadas que serão transferidas para a GPU. Aumentar esse valor pode melhorar significativamente o desempenho de modelos otimizados para aceleração de GPU, mas também pode consumir mais energia e recursos da GPU.",
 	"Set the number of worker threads used for computation. This option controls how many threads are used to process incoming requests concurrently. Increasing this value can improve performance under high concurrency workloads but may also consume more CPU resources.": "Defina o número de threads de trabalho usadas para computação. Esta opção controla quantos threads são usados para processar as solicitações recebidas de forma simultânea. Aumentar esse valor pode melhorar o desempenho em cargas de trabalho de alta concorrência, mas também pode consumir mais recursos da CPU.",
 	"Set Voice": "Definir Voz",
-	"Set web search availability for new chats.": "",
+	"Set web search availability for new chats.": "Defina a disponibilidade da pesquisa na web para novos chats.",
 	"Set whisper model": "Definir modelo Whisper",
 	"Set your status": "Defina seu status",
 	"Sets a flat bias against tokens that have appeared at least once. A higher value (e.g., 1.5) will penalize repetitions more strongly, while a lower value (e.g., 0.9) will be more lenient. At 0, it is disabled.": "Define um viés fixo contra tokens que apareceram pelo menos uma vez. Um valor mais alto (por exemplo, 1,5) penalizará as repetições com mais força, enquanto um valor mais baixo (por exemplo, 0,9) será mais tolerante. Em 0, está desabilitado.",
@@ -2393,40 +2393,40 @@
 	"Settings": "Configurações",
 	"Settings Permissions": "Configurações de Permissões",
 	"Settings saved successfully!": "Configurações salvas com sucesso!",
-	"Shapes the short label generated for each chat.": "",
+	"Shapes the short label generated for each chat.": "Define o rótulo curto gerado para cada chat.",
 	"Share": "Compartilhar",
 	"Share Chat": "Compartilhar Chat",
 	"Share link copied to clipboard.": "Link de compartilhamento copiado para a área de transferência.",
 	"Share to Open WebUI Community": "Compartilhar com a Comunidade OpenWebUI",
 	"Share your background and interests": "Fale sobre você e seus interesses",
-	"Share your current location with features that can use it.": "",
+	"Share your current location with features that can use it.": "Compartilhe sua localização atual com os recursos que podem usá-la.",
 	"Shared": "Compartilhado",
 	"Shared Chats": "Chats compartilhados",
-	"Shared prompt for retrieval and web-search query rewriting.": "",
+	"Shared prompt for retrieval and web-search query rewriting.": "Prompt compartilhado para reescrita de consultas de recuperação e pesquisa na web.",
 	"Shared with you": "Compartilhado com você",
 	"Sharing Permissions": "Permissões de compartilhamento",
 	"Show": "Mostrar",
 	"Show \"What's New\" Modal on Login": "Mostrar \"O que há de Novo\" no login",
-	"Show admin contact details while an account waits for approval.": "",
+	"Show admin contact details while an account waits for approval.": "Mostre os dados de contato do administrador enquanto uma conta aguarda aprovação.",
 	"Show All": "Mostrar Tudo",
 	"Show all ({{COUNT}} characters)": "Mostrar todos os ({{COUNT}} caracteres)",
-	"Show emoji feedback in the call interface.": "",
+	"Show emoji feedback in the call interface.": "Mostre feedback com emojis na interface de chamada.",
 	"Show Files": "Mostrar arquivos",
 	"Show Files on Terminal Select": "Mostrar Arquivos ao Selecionar Terminal",
-	"Show formatting controls in the rich text composer.": "",
+	"Show formatting controls in the rich text composer.": "Mostre os controles de formatação no editor de texto rico.",
 	"Show Formatting Toolbar": "Mostrar barra de ferramentas de formatação",
 	"Show image preview": "Mostrar pré-visualização da imagem",
 	"Show Model": "Mostrar modelo",
-	"Show or hide custom generation parameters.": "",
-	"Show or hide sensitive account secrets.": "",
+	"Show or hide custom generation parameters.": "Mostre ou oculte os parâmetros de geração personalizados.",
+	"Show or hide sensitive account secrets.": "Mostre ou oculte segredos confidenciais da conta.",
 	"Show Shortcuts": "Mostrar Atalhos",
-	"Show suggested next questions after assistant responses.": "",
-	"Show the floating quick-action toolbar in chat.": "",
-	"Show the regenerate action menu for assistant responses.": "",
-	"Show update toasts to admins when new versions are available.": "",
-	"Show user status information in the app.": "",
+	"Show suggested next questions after assistant responses.": "Mostre sugestões de próximas perguntas após as respostas do assistente.",
+	"Show the floating quick-action toolbar in chat.": "Mostre a barra de ferramentas flutuante de ações rápidas no chat.",
+	"Show the regenerate action menu for assistant responses.": "Mostre o menu da ação de regenerar nas respostas do assistente.",
+	"Show update toasts to admins when new versions are available.": "Mostre notificações de atualização aos administradores quando novas versões estiverem disponíveis.",
+	"Show user status information in the app.": "Mostre informações de status dos usuários no aplicativo.",
 	"Show your support!": "Mostre seu apoio!",
-	"Show your username label instead of You in chat bubbles.": "",
+	"Show your username label instead of You in chat bubbles.": "Mostre seu nome de usuário em vez de Você nos balões do chat.",
 	"Showcased creativity": "Criatividade exibida",
 	"Showing all messages (user + assistant) per user.": "Exibindo todas as mensagens (usuário + assistente) por usuário.",
 	"Sign in": "Entrar",
@@ -2436,7 +2436,7 @@
 	"Sign up": "Inscrever-se",
 	"Sign up to {{WEBUI_NAME}}": "Inscreva-se em {{WEBUI_NAME}}",
 	"Signing in to {{WEBUI_NAME}}": "Fazendo login em {{WEBUI_NAME}}",
-	"simultaneous sub-agents": "",
+	"simultaneous sub-agents": "subagentes simultâneos",
 	"Single": "Único",
 	"Sink List": "Lista de destino",
 	"sk-1234": "sk-1234",
@@ -2446,7 +2446,7 @@
 	"Skill ID": "ID da skill",
 	"Skill imported successfully": "Skill importada com sucesso",
 	"Skill Instructions": "Instruções da Skill",
-	"Skill Menu": "",
+	"Skill Menu": "Menu de habilidades",
 	"Skill Name": "Nome da skill",
 	"Skill updated successfully": "Skill atualizada com sucesso",
 	"Skills": "Skills",
@@ -2454,7 +2454,7 @@
 	"Skills Public Sharing": "Compartilhamento público de skills",
 	"Skills Sharing": "Compartilhamento de skills",
 	"Skip Cache": "Pular cache",
-	"Skip cached Marker results and rerun inference.": "",
+	"Skip cached Marker results and rerun inference.": "Ignore os resultados do Marker em cache e execute a inferência novamente.",
 	"Something went wrong :/": "Algo deu errado :/",
 	"Sonar": "Sonar",
 	"Sonar Deep Research": "Sonar Deep Research",
@@ -2471,10 +2471,10 @@
 	"Speech-to-Text": "Fala-para-Texto",
 	"Speech-to-Text Engine": "Motor de Transcrição de Fala",
 	"Speech-to-Text Language": "Idioma de Fala para Texto",
-	"Split documents by markdown headers before character or token splitting.": "",
-	"Sponsor on GitHub": "",
+	"Split documents by markdown headers before character or token splitting.": "Divida os documentos por cabeçalhos Markdown antes da divisão por caracteres ou tokens.",
+	"Sponsor on GitHub": "Patrocinar no GitHub",
 	"Start a new conversation": "Iniciar uma nova conversa",
-	"Start new chats as temporary unless changed.": "",
+	"Start new chats as temporary unless changed.": "Inicie novos chats como temporários, a menos que isso seja alterado.",
 	"Start of the channel": "Início do canal",
 	"Start Tag": "Tag inicial",
 	"Starting in {{count}} minutes_one": "Começando em {{count}} minuto",
@@ -2504,7 +2504,7 @@
 	"Stylized PDF Export": "Exportação de PDF estilizado",
 	"Su_day_of_week": "Dom",
 	"Sub Claim": "Reivindicação Sub",
-	"Sub-agents": "",
+	"Sub-agents": "Subagentes",
 	"Submit question": "Enviar pergunta",
 	"Submit suggestion": "Enviar sugestão",
 	"Subtitle": "Subtítulo",
@@ -2512,12 +2512,12 @@
 	"Successfully imported {{userCount}} users.": "{{userCount}} usuários importados com sucesso.",
 	"Successfully updated.": "Atualizado com sucesso.",
 	"Suggest a change": "Sugira uma mudança",
-	"Suggest completions while composing prompts.": "",
-	"Suggest completions while users type chat messages.": "",
+	"Suggest completions while composing prompts.": "Sugira conclusões durante a elaboração de prompts.",
+	"Suggest completions while users type chat messages.": "Sugira conclusões enquanto os usuários digitam mensagens no chat.",
 	"Suggested": "Sugerido",
-	"Suggested prompts": "",
-	"Summarize older chat history when the conversation context grows large.": "",
-	"Summarize this note.": "",
+	"Suggested prompts": "Prompts sugeridos",
+	"Summarize older chat history when the conversation context grows large.": "Resuma o histórico mais antigo do chat quando o contexto da conversa ficar extenso.",
+	"Summarize this note.": "Resuma esta nota.",
 	"Support": "Suporte",
 	"Support this plugin:": "Apoie este plugin:",
 	"Supported MIME Types": "Tipos MIME suportados",
@@ -2536,7 +2536,7 @@
 	"System": "Sistema",
 	"System events only": "Apenas eventos do sistema",
 	"System Instructions": "Instruções do sistema",
-	"System prompt": "",
+	"System prompt": "Prompt do sistema",
 	"System Prompt": "Prompt do Sistema",
 	"Table": "Tabela",
 	"Tag": "Tag",
@@ -2546,8 +2546,8 @@
 	"Tail free sampling is used to reduce the impact of less probable tokens from the output. A higher value (e.g., 2.0) will reduce the impact more, while a value of 1.0 disables this setting.": "A amostragem livre de cauda é usada para reduzir o impacto de tokens menos prováveis na saída. Um valor mais alto (por exemplo, 2,0) reduzirá ainda mais o impacto, enquanto um valor de 1,0 desabilita essa configuração.",
 	"Talk to Model": "Fale com o modelo",
 	"Tap to interrupt": "Toque para interromper",
-	"Target ID": "",
-	"Target ID for notify": "",
+	"Target ID": "ID do destino",
+	"Target ID for notify": "ID do destino para notify",
 	"Task List": "Lista de tarefas",
 	"Task Management": "Gerenciamento de Tarefas",
 	"Task Model": "Modelo de Tarefa",
@@ -2562,14 +2562,14 @@
 	"Terminal": "Terminal",
 	"Terminal servers saved": "Servidores de terminal salvos",
 	"Test": "Testar",
-	"Test must pass before a source is created.": "",
-	"Test notification sent.": "",
+	"Test must pass before a source is created.": "O teste deve ser aprovado antes que uma fonte seja criada.",
+	"Test notification sent.": "Notificação de teste enviada.",
 	"Test Query": "Consulta de Teste",
 	"Test returned no results.": "O teste não retornou resultados.",
 	"Test succeeded.": "Teste bem-sucedido.",
 	"Test the source before creating it.": "Teste a fonte antes de criá-la.",
 	"Test the source before saving it.": "Teste a fonte antes de salvá-la.",
-	"Text output format returned by Marker.": "",
+	"Text output format returned by Marker.": "Formato da saída de texto retornada pelo Marker.",
 	"Text Splitter": "Divisor de Texto",
 	"Text-to-Speech": "Texto-para-Fala",
 	"Text-to-Speech Engine": "Motor de Texto para Fala",
@@ -2584,7 +2584,7 @@
 	"The language of the input audio. Supplying the input language in ISO-639-1 (e.g. en) format will improve accuracy and latency. Leave blank to automatically detect the language.": "O idioma do áudio de entrada. Fornecer o idioma de entrada no formato ISO-639-1 (por exemplo, en) aumentará a precisão e a latência. Deixe em branco para detectar o idioma automaticamente.",
 	"The LDAP attribute that maps to the mail that users use to sign in.": "O atributo LDAP que mapeia o e-mail que os usuários usam para fazer login.",
 	"The LDAP attribute that maps to the username that users use to sign in.": "O atributo LDAP que mapeia para o nome de usuário que os usuários usam para fazer login.",
-	"The notify tool always sends to an enabled target, regardless of automatic event settings.": "",
+	"The notify tool always sends to an enabled target, regardless of automatic event settings.": "A ferramenta notify sempre envia para um destino habilitado, independentemente das configurações de eventos automáticos.",
 	"The passwords you entered don't quite match. Please double-check and try again.": "As senhas digitadas não coincidem. Por favor, verifique e tente novamente.",
 	"The score should be a value between 0.0 (0%) and 1.0 (100%).": "A pontuação deve ser um valor entre 0.0 (0%) e 1.0 (100%).",
 	"The stream delta chunk size for the model. Increasing the chunk size will make the model respond with larger pieces of text at once.": "O tamanho do bloco delta do fluxo para o modelo. Aumentar o tamanho do bloco fará com que o modelo responda com trechos maiores de texto de uma só vez.",
@@ -2619,17 +2619,17 @@
 	"Thought for {{DURATION}} seconds": "Pensado por {{DURATION}} segundos",
 	"Thought for less than a second": "Pensou por menos de um segundo",
 	"Thread": "Tópico",
-	"Through": "",
+	"Through": "Até",
 	"Thumbs up/down ratings from users on model responses": "Avaliações positivas/negativas dos usuários sobre as respostas do modelo",
 	"Tika": "Tika",
-	"Tika server endpoint used for content extraction.": "",
-	"Tika Server URL": "",
+	"Tika server endpoint used for content extraction.": "Endpoint do servidor Tika usado para extração de conteúdo.",
+	"Tika Server URL": "URL do servidor Tika",
 	"Tika Server URL required.": "URL do servidor Tika necessária.",
 	"Tiktoken": "Tiktoken",
 	"Time": "Tempo",
 	"Time & Calculation": "Tempo e Cálculo",
 	"Timeout": "Tempo limite",
-	"Timer": "",
+	"Timer": "Temporizador",
 	"Title": "Título",
 	"Title Auto-Generation": "Geração Automática de Título",
 	"Title cannot be an empty string.": "O Título não pode ser uma string vazia.",
@@ -2637,19 +2637,19 @@
 	"Title Generation Prompt": "Prompt de Geração de Título",
 	"Title is required": "O título é obrigatório",
 	"TLS": "TLS",
-	"TLS cipher list for LDAP connections.": "",
+	"TLS cipher list for LDAP connections.": "Lista de cifras TLS para conexões LDAP.",
 	"To access the available model names for downloading,": "Para acessar os nomes de modelos disponíveis para download,",
 	"To access the GGUF models available for downloading,": "Para acessar os modelos GGUF disponíveis para download,",
 	"To access the WebUI, please reach out to the administrator. Admins can manage user statuses from the Admin Panel.": "Para acessar a WebUI, entre em contato com o administrador. Os administradores podem gerenciar os status dos usuários no Painel de Administração.",
 	"To attach knowledge base here, add them to the \"Knowledge\" workspace first.": "Para anexar a base de conhecimento aqui, adicione-os ao espaço de trabalho \"Conhecimento\" primeiro.",
 	"To learn more about available endpoints, visit our documentation.": "Para saber mais sobre os endpoints disponíveis, visite nossa documentação.",
-	"To learn more about SpeechT5,": "",
-	"To see the available CMU Arctic speaker embeddings,": "",
-	"To select actions here, add them to the \"Functions\" workspace first.": "",
-	"To select default filters here, enable toggleable filters for this model first.": "",
-	"To select filters here, add them to the \"Functions\" workspace first.": "",
+	"To learn more about SpeechT5,": "Para saber mais sobre o SpeechT5,",
+	"To see the available CMU Arctic speaker embeddings,": "Para ver os embeddings de voz CMU Arctic disponíveis,",
+	"To select actions here, add them to the \"Functions\" workspace first.": "Para selecionar ações aqui, primeiro adicione-as ao workspace de \"Funções\".",
+	"To select default filters here, enable toggleable filters for this model first.": "Para selecionar filtros padrão aqui, primeiro habilite filtros alternáveis para este modelo.",
+	"To select filters here, add them to the \"Functions\" workspace first.": "Para selecionar filtros aqui, primeiro adicione-os ao workspace de \"Funções\".",
 	"To select skills here, add them to the \"Skills\" workspace first.": "Para selecionar skills aqui, adicione-as primeiro ao espaço de trabalho \"Skills\".",
-	"To select toolkits here, add them to the \"Tools\" workspace or enable a tool server first.": "",
+	"To select toolkits here, add them to the \"Tools\" workspace or enable a tool server first.": "Para selecionar kits de ferramentas aqui, primeiro adicione-os ao workspace de \"Ferramentas\" ou habilite um servidor de ferramentas.",
 	"Toast Notifications for New Updates": "Notificações de alerta para novas atualizações",
 	"Today": "Hoje",
 	"Today at": "Hoje em",
@@ -2662,12 +2662,12 @@
 	"Toggle status history": "Alternar histórico de status",
 	"Toggle whether current connection is active.": "Alterna se a conexão atual está ativa.",
 	"Token": "Token",
-	"Token activity": "",
-	"Token Cap": "",
+	"Token activity": "Atividade de tokens",
+	"Token Cap": "Limite de tokens",
 	"Token counts are estimates and may not reflect actual API usage": "A contagem de tokens é uma estimativa e pode não refletir o uso real da API.",
 	"Token Threshold": "Limite de Token",
 	"Tokenizer Model": "Modelo de Tokenizador",
-	"Tokenizer model used for transformer token splitting.": "",
+	"Tokenizer model used for transformer token splitting.": "Modelo de tokenizador usado para a divisão de tokens do transformer.",
 	"tokens": "tokens",
 	"Tokens": "Tokens",
 	"Too verbose": "Muito detalhado",
@@ -2676,14 +2676,14 @@
 	"Tool Description": "Descrição da ferramenta",
 	"Tool ID": "ID da ferramenta",
 	"Tool imported successfully": "Ferramenta importada com sucesso",
-	"tool loops per sub-agent": "",
-	"Tool Menu": "",
+	"tool loops per sub-agent": "ciclos de ferramentas por subagente",
+	"Tool Menu": "Menu de ferramentas",
 	"Tool Name": "Nome da ferramenta",
 	"Tool Servers": "Servidores de ferramentas",
 	"Tool updated successfully": "Ferramenta atualizada com sucesso",
 	"Tools": "Ferramentas",
 	"Tools Access": "Acesso às Ferramentas",
-	"Tools can execute arbitrary code.": "",
+	"Tools can execute arbitrary code.": "As ferramentas podem executar código arbitrário.",
 	"Tools Function Calling Prompt": "Prompt de chamada de função de ferramentas",
 	"Tools have a function calling system that allows arbitrary code execution.": "Ferramentas possuem um sistema de chamada de funções que permite a execução de código arbitrário.",
 	"Tools Public Sharing": "Compartilhamento Público de Ferramentas",
@@ -2691,8 +2691,8 @@
 	"Top": "Top",
 	"Top K": "Top K",
 	"Top K Reranker": "Top K Reclassificador",
-	"Top models": "",
-	"Total chats": "",
+	"Top models": "Principais modelos",
+	"Total chats": "Total de chats",
 	"Transformers": "Transformers",
 	"Trouble accessing Ollama?": "Problemas para acessar o Ollama?",
 	"Trust Proxy Environment": "Ambiente de Proxy Confiável",
@@ -2711,9 +2711,9 @@
 	"Unarchive All": "Desarquivar tudo",
 	"Unarchive All Archived Chats": "Desarquivar Todos os Chats Arquivados",
 	"Unarchive Chat": "Desarquivar Chat",
-	"Unassigned": "",
+	"Unassigned": "Não atribuído",
 	"Underline": "Sublinhado",
-	"unknown": "",
+	"unknown": "desconhecido",
 	"Unknown": "Desconhecido",
 	"Unknown User": "Usuário desconhecido",
 	"Unloads {{FROM_NOW}}": "Descarrega {{FROM_NOW}}",
@@ -2729,13 +2729,13 @@
 	"Update": "Atualizar",
 	"Update and Copy Link": "Atualizar e Copiar Link",
 	"Update Email": "Atualizar Email",
-	"Update embedding model": "",
+	"Update embedding model": "Atualizar modelo de embedding",
 	"Update for the latest features and improvements.": "Atualizar para as novas funcionalidades e melhorias.",
-	"Update model": "",
+	"Update model": "Atualizar modelo",
 	"Update Name": "Atualizar Nome",
 	"Update password": "Atualizar senha",
 	"Update Picture": "Atualizar Foto",
-	"Update the password used for email and password sign-in.": "",
+	"Update the password used for email and password sign-in.": "Atualize a senha usada para entrar com e-mail e senha.",
 	"Update your status": "Atualize seu status",
 	"Updated": "Atualizado",
 	"Updated at": "Atualizado em",
@@ -2743,20 +2743,20 @@
 	"Upgrade to a licensed plan for enhanced capabilities, including custom theming and branding, and dedicated support.": "Faça upgrade para um plano licenciado para obter recursos aprimorados, incluindo temas e marcas personalizados e suporte dedicado.",
 	"Upload": "Fazer upload",
 	"Upload a GGUF model": "Fazer upload de um modelo GGUF",
-	"Upload a local Python pipeline file to the selected server.": "",
-	"Upload a workflow.json file exported as API format from ComfyUI.": "",
+	"Upload a local Python pipeline file to the selected server.": "Envie um arquivo de pipeline Python local ao servidor selecionado.",
+	"Upload a workflow.json file exported as API format from ComfyUI.": "Envie um arquivo workflow.json exportado do ComfyUI no formato de API.",
 	"Upload Audio": "Carregar Áudio",
 	"Upload directory": "Carregar diretório",
 	"Upload files": "Carregar arquivos",
 	"Upload Files": "Carregar Arquivos",
 	"Upload Model": "Carregar Modelo",
-	"Upload or reset the image shown behind chat content.": "",
-	"Upload Photo": "",
+	"Upload or reset the image shown behind chat content.": "Envie ou redefina a imagem exibida atrás do conteúdo do chat.",
+	"Upload Photo": "Enviar foto",
 	"Upload Pipeline": "Fazer upload de Pipeline",
 	"Upload profile image": "Enviar imagem de perfil",
 	"Upload Progress": "Progresso do Upload",
 	"Uploaded files or images": "Arquivos ou imagens carregados",
-	"Uploading": "",
+	"Uploading": "Enviando",
 	"Uploading {{current}}/{{total}}: {{file}}": "Enviando {{current}}/{{total}}: {{file}}",
 	"Uploading...": "Enviando...",
 	"URL": "URL",
@@ -2765,36 +2765,36 @@
 	"Usage": "Uso",
 	"Use": "Usar",
 	"Use /v1/chat/completions endpoint instead of /v1/audio/transcriptions for potentially better accuracy.": "Use o endpoint /v1/chat/completions em vez de /v1/audio/transcriptions para obter uma precisão potencialmente melhor.",
-	"Use a Gemini API key for image generation.": "",
+	"Use a Gemini API key for image generation.": "Use uma chave de API do Gemini para geração de imagens.",
 	"Use a valid event name or pattern like user.*": "Use um nome de evento válido ou padrão como user.*",
-	"Use a wider chat layout on large displays.": "",
-	"Use an API key when your ComfyUI server requires one.": "",
-	"Use an LLM to improve tables, forms, math, and layout detection.": "",
+	"Use a wider chat layout on large displays.": "Use um layout de chat mais amplo em telas grandes.",
+	"Use an API key when your ComfyUI server requires one.": "Use uma chave de API quando seu servidor ComfyUI exigir uma.",
+	"Use an LLM to improve tables, forms, math, and layout detection.": "Use um LLM para melhorar a detecção de tabelas, formulários, expressões matemáticas e layout.",
 	"Use Base64": "Usar Base64",
 	"Use Chat Completions API": "Usar API de Chat Completions",
-	"Use device vibration feedback on supported Android devices.": "",
+	"Use device vibration feedback on supported Android devices.": "Use o feedback por vibração do dispositivo em aparelhos Android compatíveis.",
 	"Use discovered scopes": "Usar escopos descobertos",
 	"Use groups to organize your users and assign permissions.": "Use grupos para organizar seus usuários e atribuir permissões.",
 	"Use LLM": "Usar LLM",
-	"Use proxy environment variables to fetch page contents.": "",
-	"Use search results without fetching page contents.": "",
-	"Use segmented retrieval for focused and relevant content extraction.": "",
-	"Use segmented retrieval for focused and relevant context.": "",
-	"Use segmented retrieval for focused context.": "",
-	"Use styled formatting when exporting chats to PDF.": "",
-	"Use the active chat title as the browser tab title.": "",
-	"Use the rich composer instead of a plain textarea.": "",
-	"Use TLS when connecting to the LDAP server.": "",
+	"Use proxy environment variables to fetch page contents.": "Use variáveis de ambiente de proxy para obter o conteúdo das páginas.",
+	"Use search results without fetching page contents.": "Use os resultados da pesquisa sem obter o conteúdo das páginas.",
+	"Use segmented retrieval for focused and relevant content extraction.": "Use a recuperação segmentada para extrair conteúdo relevante e específico.",
+	"Use segmented retrieval for focused and relevant context.": "Use a recuperação segmentada para obter um contexto relevante e específico.",
+	"Use segmented retrieval for focused context.": "Use a recuperação segmentada para obter um contexto específico.",
+	"Use styled formatting when exporting chats to PDF.": "Use formatação estilizada ao exportar chats para PDF.",
+	"Use the active chat title as the browser tab title.": "Use o título do chat ativo como título da aba do navegador.",
+	"Use the rich composer instead of a plain textarea.": "Use o editor de texto rico em vez de uma área de texto simples.",
+	"Use TLS when connecting to the LDAP server.": "Use TLS ao se conectar ao servidor LDAP.",
 	"Use Web Search?": "Usar Pesquisa na Web?",
 	"user": "usuário",
 	"User": "Usuário",
 	"User Access": "Acesso do Usuário",
 	"User Activity": "Atividade do usuário",
-	"User Connections": "",
+	"User Connections": "Conexões do usuário",
 	"User Groups": "Grupos de usuários",
 	"User location successfully retrieved.": "Localização do usuário recuperada com sucesso.",
 	"User menu": "Menu do usuário",
-	"User messages": "",
+	"User messages": "Mensagens do usuário",
 	"User Preview": "Visualização do Usuário",
 	"User ratings (thumbs up/down)": "Avaliações dos usuários (polegar para cima/polegar para baixo)",
 	"User Status": "Status do usuário",
@@ -2804,14 +2804,14 @@
 	"users": "usuários",
 	"Users": "Usuários",
 	"Uses DefaultAzureCredential to authenticate": "Usa DefaultAzureCredential para autenticar",
-	"Uses defaults": "",
+	"Uses defaults": "Usa os padrões",
 	"Uses OAuth 2.1 Dynamic Client Registration": "Utiliza o registro dinâmico de cliente OAuth 2.1",
 	"Using Entire Document": "Usando o documento inteiro",
 	"Using Focused Retrieval": "Usando Recuperação Focada",
 	"Using the default arena model with all models. Click the plus button to add custom models.": "Usando a arena de modelos padrão para todos os modelos. Clique no botão mais para adicionar modelos personalizados.",
-	"Valid time units: 's', 'm', 'h', 'd', 'w' or '-1' for no expiration.": "",
-	"Validate Certificate": "",
-	"Validate SSL certificates when fetching web content.": "",
+	"Valid time units: 's', 'm', 'h', 'd', 'w' or '-1' for no expiration.": "Unidades de tempo válidas: 's', 'm', 'h', 'd', 'w' ou '-1' para não expirar.",
+	"Validate Certificate": "Validar certificado",
+	"Validate SSL certificates when fetching web content.": "Valide os certificados SSL ao obter conteúdo da web.",
 	"Valves": "Configurações",
 	"Valves updated": "Configurações atualizadas",
 	"Valves updated successfully": "Configurações atualizadas com sucesso",
@@ -2819,16 +2819,16 @@
 	"Vector Field": "Campo de Vetor",
 	"Verify Connection": "Verificar conexão",
 	"Verify SSL Certificate": "Verificar certificado SSL",
-	"Verify the LDAP server certificate when TLS is enabled.": "",
+	"Verify the LDAP server certificate when TLS is enabled.": "Verifique o certificado do servidor LDAP quando o TLS estiver habilitado.",
 	"Version": "Versão",
 	"Version {{selectedVersion}} of {{totalVersions}}": "Versão {{selectedVersion}} de {{totalVersions}}",
 	"Version deleted": "Versão excluída",
-	"View chat": "",
+	"View chat": "Ver chat",
 	"View Replies": "Ver respostas",
-	"View Result from {{NAME}}": "",
+	"View Result from {{NAME}}": "Ver resultado de {{NAME}}",
 	"View source: {{name}}": "Ver origem: {{name}}",
 	"View source: {{title}}": "Ver origem: {{title}}",
-	"View the installed version and check release updates.": "",
+	"View the installed version and check release updates.": "Veja a versão instalada e verifique as atualizações de versão.",
 	"Visibility": "Visibilidade",
 	"Visible": "Visível",
 	"Visible to all users": "Visível para todos os usuários",
@@ -2838,8 +2838,8 @@
 	"Voice Input": "Entrada de voz",
 	"Voice mode": "Modo de voz",
 	"Voice Mode Prompt": "Prompt do Modo de Voz",
-	"Wait for the current response to finish before compacting.": "",
-	"Wait for the current response to finish before forking.": "",
+	"Wait for the current response to finish before compacting.": "Aguarde a conclusão da resposta atual antes de compactar.",
+	"Wait for the current response to finish before forking.": "Aguarde a conclusão da resposta atual antes de bifurcar.",
 	"Waiting for upload...": "Aguardando upload...",
 	"Warning": "Aviso",
 	"Warning:": "Aviso:",
@@ -2856,7 +2856,7 @@
 	"Web Search Engine": "Mecanismo de Busca na Web",
 	"Web Search in Chat": "Pesquisa na Web no Chat",
 	"Web Search Query Generation": "Geração de consulta de pesquisa na Web",
-	"Webhook": "",
+	"Webhook": "Webhook",
 	"Webhook deleted": "Webhook excluído",
 	"Webhook Name": "Nome do Webhook",
 	"Webhook saved": "Webhook salvo",
@@ -2868,10 +2868,10 @@
 	"WebUI will make requests to \"{{url}}/api/chat\"": "A WebUI fará requisições para \"{{url}}/api/chat\".",
 	"WebUI will make requests to \"{{url}}/chat/completions\"": "A WebUI fará requisições para \"{{url}}/chat/completions\".",
 	"Week": "Semana",
-	"Week of": "",
+	"Week of": "Semana de",
 	"Weekly": "Semanal",
-	"Weekly at": "",
-	"Welcome to your AI home.": "",
+	"Weekly at": "Semanalmente às",
+	"Welcome to your AI home.": "Boas-vindas à sua central de IA.",
 	"What are you trying to achieve?": "O que está tentando alcançar?",
 	"What are you working on?": "No que está trabalhando?",
 	"What is NOT shared:": "O que NÃO é compartilhado:",
@@ -2885,7 +2885,7 @@
 	"Why?": "Por que?",
 	"Widescreen Mode": "Modo Widescreen",
 	"Width": "Largura",
-	"Width in pixels to compress images to. Leave empty for no compression.": "",
+	"Width in pixels to compress images to. Leave empty for no compression.": "Largura em pixels para a qual as imagens serão comprimidas. Deixe vazio para não comprimir.",
 	"Wikipedia": "Wikipédia",
 	"Won": "Ganhou",
 	"Working Directory": "Diretório de Trabalho",
@@ -2907,7 +2907,7 @@
 	"Yearly": "Anual",
 	"Yesterday": "Ontem",
 	"You": "Você",
-	"You are a sub-agent...": "",
+	"You are a sub-agent...": "Você é um subagente...",
 	"You are currently using a trial license. Please contact support to upgrade your license.": "Você está usando uma licença de teste. Entre em contato com o suporte para atualizar sua licença.",
 	"You can only chat with a maximum of {{maxCount}} file(s) at a time.": "Você só pode conversar com no máximo {{maxCount}} arquivo(s) de cada vez.",
 	"You can personalize your interactions with LLMs by adding memories through the 'Manage' button below, making them more helpful and tailored to you.": "Você pode personalizar suas interações com LLMs adicionando memórias através do botão 'Gerenciar' abaixo, tornando-as mais úteis e adaptadas a você.",
@@ -2922,7 +2922,7 @@
 	"You do not have permission to upload files to this knowledge base.": "Você não tem permissão para enviar arquivos para esta base de conhecimento.",
 	"You do not have permission to upload files.": "Você não tem permissão para fazer upload de arquivos.",
 	"You do not have permission to upload web content.": "Você não tem permissão para enviar conteúdo da web.",
-	"You have more than 50 users, which often means this workspace is supporting organizational use. Open WebUI is free to use as-is, with no restrictions or hidden limits, and we want to keep it that way.": "",
+	"You have more than 50 users, which often means this workspace is supporting organizational use. Open WebUI is free to use as-is, with no restrictions or hidden limits, and we want to keep it that way.": "Você tem mais de 50 usuários, o que geralmente significa que este workspace atende a uma organização. O Open WebUI é gratuito para uso como está, sem restrições ou limites ocultos, e queremos mantê-lo assim.",
 	"You have no archived conversations.": "Você não tem conversas arquivadas.",
 	"You have no shared conversations.": "Você não tem conversas compartilhadas.",
 	"You have shared this chat": "Você compartilhou este chat",

From adf07e8df0a2b9eb592f9cec64476037a5e0a909 Mon Sep 17 00:00:00 2001
From: Andi Chandler <andi@gowling.com>
Date: Thu, 23 Jul 2026 08:24:21 +0100
Subject: [PATCH 12/29] **i18n**: Internationalization or localization changes
 (#27304)

* Change spelling to British English in translation.json

Updated translations to use British English spelling for various terms.

* Update translation.json
---
 src/lib/i18n/locales/en-GB/translation.json | 44 ++++++++++-----------
 1 file changed, 22 insertions(+), 22 deletions(-)

diff --git a/src/lib/i18n/locales/en-GB/translation.json b/src/lib/i18n/locales/en-GB/translation.json
index d2ad49ae6a..5df7fa1c8a 100644
--- a/src/lib/i18n/locales/en-GB/translation.json
+++ b/src/lib/i18n/locales/en-GB/translation.json
@@ -38,7 +38,7 @@
 	"{{COUNT}}y_time_ago": "",
 	"{{label}} contains invalid JSON": "",
 	"{{label}} must be a JSON object": "",
-	"{{model}} download has been canceled": "",
+	"{{model}} download has been canceled": "{{model}} download has been cancelled",
 	"{{modelName}} profile image": "",
 	"{{NAMES}} reacted with {{REACTION}}": "",
 	"{{user}}'s Chats": "",
@@ -198,8 +198,8 @@
 	"Allow users to create API keys for programmatic access.": "",
 	"Allow users to edit existing images.": "",
 	"Allow users to generate images from prompts.": "",
-	"Allow users to organize chats into folders.": "",
-	"Allow users to save memories for more personalized responses.": "",
+	"Allow users to organize chats into folders.": "Allow users to organise chats into folders.",
+	"Allow users to save memories for more personalized responses.": "Allow users to save memories for more personalised responses.",
 	"Allow users to search the web from chats.": "",
 	"Allow users to share chats with the Open WebUI community.": "",
 	"Allow users to use channels for shared conversations.": "",
@@ -377,7 +377,7 @@
 	"Button Prompt": "",
 	"by {{name}}": "",
 	"By {{name}}": "",
-	"By supporting the project through sponsorship or an enterprise license, you help us stay independent, ship new features faster, improve stability, and grow Open WebUI for the long haul.": "",
+	"By supporting the project through sponsorship or an enterprise license, you help us stay independent, ship new features faster, improve stability, and grow Open WebUI for the long haul.": "By supporting the project through sponsorship or an enterprise licence, you help us stay independent, ship new features faster, improve stability, and grow Open WebUI for the long haul.",
 	"Bypass Embedding and Retrieval": "",
 	"Bypass Web Loader": "",
 	"Cache Base Model List": "",
@@ -444,7 +444,7 @@
 	"Choose how uploaded documents are parsed before indexing.": "",
 	"Choose how web result pages are fetched and read.": "",
 	"Choose the browser voice used for speech output.": "",
-	"Choose the color theme used by the interface.": "",
+	"Choose the color theme used by the interface.": "Choose the colour theme used by the interface.",
 	"Choose the configured text-to-speech service voice.": "",
 	"Choose the engine used to read assistant responses aloud.": "",
 	"Choose the engine used to transcribe voice input.": "",
@@ -676,7 +676,7 @@
 	"Current Password": "",
 	"Current streak": "",
 	"Custom": "",
-	"Custom color": "",
+	"Custom color": "Custom colour",
 	"Custom description enabled": "",
 	"Custom Gender": "",
 	"Custom message shown while an account waits for approval.": "",
@@ -835,7 +835,7 @@
 	"Download all users as CSV.": "",
 	"Download as JSON": "",
 	"Download as SVG": "",
-	"Download canceled": "",
+	"Download canceled": "Download cancelled",
 	"Download every user's chat history as JSON.": "",
 	"Download the application database when supported.": "",
 	"Download the current admin configuration as JSON.": "",
@@ -914,7 +914,7 @@
 	"Enable Mirostat sampling for controlling perplexity.": "",
 	"Enable sub-agents": "",
 	"Enable Web Search": "",
-	"Enable, disable, or customize the reasoning tags used by the model. \"Enabled\" uses default tags, \"Disabled\" turns off reasoning tags, and \"Custom\" lets you specify your own start and end tags.": "",
+	"Enable, disable, or customize the reasoning tags used by the model. \"Enabled\" uses default tags, \"Disabled\" turns off reasoning tags, and \"Custom\" lets you specify your own start and end tags.": "Enable, disable, or customise the reasoning tags used by the model. \"Enabled\" uses default tags, \"Disabled\" turns off reasoning tags, and \"Custom\" lets you specify your own start and end tags.",
 	"Enabled": "",
 	"End Tag": "",
 	"Endpoint": "",
@@ -971,7 +971,7 @@
 	"Enter GitHub Raw URL": "",
 	"Enter Google PSE API Key": "",
 	"Enter Google PSE Engine Id": "",
-	"Enter hex color (e.g. #FF0000)": "",
+	"Enter hex color (e.g. #FF0000)": "Enter hex colour (e.g. #FF0000)",
 	"Enter Image Size (e.g. 512x512)": "",
 	"Enter Jina API Base URL": "",
 	"Enter Jina API Key": "",
@@ -1034,7 +1034,7 @@
 	"Enter Tika Server URL": "",
 	"Enter to Send": "",
 	"Enter token threshold": "",
-	"Enter Tokenizer Model": "",
+	"Enter Tokenizer Model": "Enter Tokeniser Model",
 	"Enter Top K": "",
 	"Enter Top K Reranker": "",
 	"Enter URL (e.g. http://127.0.0.1:7860/)": "",
@@ -1059,7 +1059,7 @@
 	"Enter Your Password": "",
 	"Enter Your Role": "",
 	"Enter Your Username": "",
-	"Enterprise licenses also include dedicated support, customization options, and more, at a fraction of the cost of building and maintaining this stack internally.": "",
+	"Enterprise licenses also include dedicated support, customization options, and more, at a fraction of the cost of building and maintaining this stack internally.": "Enterprise licences also include dedicated support, customisation options, and more, at a fraction of the cost of building and maintaining this stack internally.",
 	"Enterprise licensing": "",
 	"Entra ID": "",
 	"Environment Variables": "",
@@ -1095,7 +1095,7 @@
 	"Example: mail": "",
 	"Example: ou=users,dc=foo,dc=example": "",
 	"Example: sAMAccountName or uid or userPrincipalName": "",
-	"Exceeded the number of seats in your license. Please contact support to increase the number of seats.": "",
+	"Exceeded the number of seats in your license. Please contact support to increase the number of seats.": "Exceeded the number of seats in your licence. Please contact support to increase the number of seats.",
 	"Exclude": "",
 	"Execute code": "",
 	"Execute code for analysis": "",
@@ -1173,7 +1173,7 @@
 	"Failed to refresh terminals: {{error}}": "",
 	"Failed to remove member": "",
 	"Failed to render diagram": "",
-	"Failed to render visualization": "",
+	"Failed to render visualization": "Failed to render visualisation",
 	"Failed to save connections": "",
 	"Failed to save conversation": "",
 	"Failed to save models configuration": "",
@@ -1523,7 +1523,7 @@
 	"Let speech interrupt the assistant during a voice call.": "",
 	"Let users rate assistant responses.": "",
 	"lexical": "",
-	"License": "",
+	"License": "Licence",
 	"Lifecycle JSON": "",
 	"Lifetime tokens": "",
 	"Lift List": "",
@@ -1859,7 +1859,7 @@
 	"October": "",
 	"Off": "",
 	"Okay, Let's Go!": "",
-	"Older messages are summarized when estimated context exceeds this token limit.": "",
+	"Older messages are summarized when estimated context exceeds this token limit.": "Older messages are summarised when estimated context exceeds this token limit.",
 	"OLED Dark": "",
 	"Ollama": "",
 	"Ollama API": "",
@@ -2502,8 +2502,8 @@
 	"Suggest completions while users type chat messages.": "",
 	"Suggested": "",
 	"Suggested prompts": "",
-	"Summarize older chat history when the conversation context grows large.": "",
-	"Summarize this note.": "",
+	"Summarize older chat history when the conversation context grows large.": "Summarise older chat history when the conversation context grows large.",
+	"Summarize this note.": "Summarise this note.",
 	"Support": "",
 	"Support this plugin:": "",
 	"Supported MIME Types": "",
@@ -2652,8 +2652,8 @@
 	"Token Cap": "",
 	"Token counts are estimates and may not reflect actual API usage": "",
 	"Token Threshold": "",
-	"Tokenizer Model": "",
-	"Tokenizer model used for transformer token splitting.": "",
+	"Tokenizer Model": "Tokeniser Model",
+	"Tokenizer model used for transformer token splitting.": "Tokeniser model used for transformer token splitting.",
 	"tokens": "",
 	"Tokens": "",
 	"Too verbose": "",
@@ -2760,7 +2760,7 @@
 	"Use Chat Completions API": "",
 	"Use device vibration feedback on supported Android devices.": "",
 	"Use discovered scopes": "",
-	"Use groups to organize your users and assign permissions.": "",
+	"Use groups to organize your users and assign permissions.": "Use groups to organise your users and assign permissions.",
 	"Use LLM": "",
 	"Use proxy environment variables to fetch page contents.": "",
 	"Use search results without fetching page contents.": "",
@@ -2894,7 +2894,7 @@
 	"Yesterday": "",
 	"You": "",
 	"You are a sub-agent...": "",
-	"You are currently using a trial license. Please contact support to upgrade your license.": "",
+	"You are currently using a trial license. Please contact support to upgrade your license.": "You are currently using a trial licence. Please contact support to upgrade your licence.",
 	"You can only chat with a maximum of {{maxCount}} file(s) at a time.": "",
 	"You can personalize your interactions with LLMs by adding memories through the 'Manage' button below, making them more helpful and tailored to you.": "You can personalise your interactions with LLMs by adding memories through the 'Manage' button below, making them more helpful and tailored to you.",
 	"You cannot upload an empty file.": "",
@@ -2908,7 +2908,7 @@
 	"You do not have permission to upload files to this knowledge base.": "",
 	"You do not have permission to upload files.": "",
 	"You do not have permission to upload web content.": "",
-	"You have more than 50 users, which often means this workspace is supporting organizational use. Open WebUI is free to use as-is, with no restrictions or hidden limits, and we want to keep it that way.": "",
+	"You have more than 50 users, which often means this workspace is supporting organizational use. Open WebUI is free to use as-is, with no restrictions or hidden limits, and we want to keep it that way.": "You have more than 50 users, which often means this workspace is supporting organisational use. Open WebUI is free to use as-is, with no restrictions or hidden limits, and we want to keep it that way.",
 	"You have no archived conversations.": "",
 	"You have no shared conversations.": "",
 	"You have shared this chat": "",

From 3cf3f8e1894d0b25e2db5aeceea22041d10e5f5b Mon Sep 17 00:00:00 2001
From: Classic298 <27028174+Classic298@users.noreply.github.com>
Date: Thu, 23 Jul 2026 09:24:51 +0200
Subject: [PATCH 13/29] fix: add missing folders field to SharingPermissions so
 the permission persists (#27296)

`sharing.folders` is present in `DEFAULT_USER_PERMISSIONS` but absent from the `SharingPermissions` response/update schema, so the admin default and group permission API silently drops it on every round-trip and the setting is never saved. Add `folders: bool = False`, matching the config default (`USER_PERMISSIONS_FOLDERS_ALLOW_SHARING`), restoring parity with `DEFAULT_USER_PERMISSIONS`.

Fixes #27120.
---
 backend/open_webui/routers/users.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/backend/open_webui/routers/users.py b/backend/open_webui/routers/users.py
index eaaa0092b2..029439ce3e 100644
--- a/backend/open_webui/routers/users.py
+++ b/backend/open_webui/routers/users.py
@@ -200,6 +200,7 @@ class SharingPermissions(BaseModel):
     public_skills: bool = False
     notes: bool = False
     public_notes: bool = True
+    folders: bool = False
     public_chats: bool = False
     public_calendars: bool = False
 

From af629177f46fa4595175f914c971c47701d1a676 Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 03:31:32 -0400
Subject: [PATCH 14/29] refac

---
 backend/open_webui/routers/users.py | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/backend/open_webui/routers/users.py b/backend/open_webui/routers/users.py
index 029439ce3e..cff598b236 100644
--- a/backend/open_webui/routers/users.py
+++ b/backend/open_webui/routers/users.py
@@ -476,6 +476,23 @@ async def update_user_settings_by_session_user(
         # If the user is not an admin and does not have permission to use tool servers, remove the key
         updated_user_settings['ui'].pop('toolServers', None)
 
+    ui_notifications = ui_settings.get('notifications') if isinstance(ui_settings, dict) else None
+    if (
+        user.role != 'admin'
+        and (
+            'notifications' in updated_user_settings
+            or (isinstance(ui_notifications, dict) and 'webhook_url' in ui_notifications)
+        )
+        and not await has_permission(
+            user.id,
+            'features.webhooks',
+            await Config.get('user.permissions'),
+        )
+    ):
+        updated_user_settings.pop('notifications', None)
+        if isinstance(ui_notifications, dict):
+            ui_notifications.pop('webhook_url', None)
+
     user = await Users.update_user_settings_by_id(user.id, updated_user_settings, db=db)
     if user:
         await publish_event(

From e18e249d5da3d8fe701a885edc64341cc5dbf813 Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 03:39:30 -0400
Subject: [PATCH 15/29] refac

---
 backend/open_webui/tools/knowledge_fs.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/backend/open_webui/tools/knowledge_fs.py b/backend/open_webui/tools/knowledge_fs.py
index 0ec1cc8983..44d092e792 100644
--- a/backend/open_webui/tools/knowledge_fs.py
+++ b/backend/open_webui/tools/knowledge_fs.py
@@ -686,12 +686,12 @@ async def _kb_grep(
 
     # Grep on piped input
     if piped_input is not None:
-        lines = piped_input.split('\\n')
+        lines = piped_input.split('\n')
         matched = []
         for i, line in enumerate(lines, 1):
             if _matches(line):
                 matched.append(f'{i}: {line}')
-        return '\\n'.join(matched) if matched else f'No matches for "{pattern}"'
+        return '\n'.join(matched) if matched else f'No matches for "{pattern}"'
 
     # Single file grep
     if file_ref and not dir_scope:
@@ -702,7 +702,7 @@ async def _kb_grep(
         elif 'error' in resolved:
             return resolved['error']
         else:
-            lines = resolved['content'].split('\\n')
+            lines = resolved['content'].split('\n')
             matched = []
             for i, line in enumerate(lines, 1):
                 if _matches(line):
@@ -715,7 +715,7 @@ async def _kb_grep(
 
             if not matched:
                 return f'No matches for "{pattern}" in {resolved["filename"]}'
-            return '\\n'.join(matched)
+            return '\n'.join(matched)
 
     # Cross-file grep (optionally scoped to directory)
     accessible = await _get_accessible_files(user, model_knowledge)

From 48cc9d388e3afbeb55b727b71a444f63c710a6b1 Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 03:39:56 -0400
Subject: [PATCH 16/29] refac

---
 backend/open_webui/functions.py | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/backend/open_webui/functions.py b/backend/open_webui/functions.py
index 6e82c0b022..1900ee752c 100644
--- a/backend/open_webui/functions.py
+++ b/backend/open_webui/functions.py
@@ -147,7 +147,10 @@ async def get_function_models(request):
     return pipe_models
 
 
-async def generate_function_chat_completion(request, form_data, user, models: dict = {}):
+async def generate_function_chat_completion(request, form_data, user, models: dict | None = None):
+    if models is None:
+        models = {}
+
     async def execute_pipe(pipe, params):
         if inspect.iscoroutinefunction(pipe):
             return await pipe(**params)

From 073cd65afe043ce73d5c12fa7a3104df712e6bff Mon Sep 17 00:00:00 2001
From: Classic298 <27028174+Classic298@users.noreply.github.com>
Date: Thu, 23 Jul 2026 09:42:48 +0200
Subject: [PATCH 17/29] Log upstream provider rejections at warn/error level
 (#27238)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When an upstream provider rejects a request (e.g. a 400 for a
max_tokens value above the model's ceiling), the actionable error
message was only published to event sinks, which are invisible unless
an event function or webhook is configured. Admins had to query the
provider's API directly to diagnose failures (open-webui#27237).

Add a single log line in publish_model_provider_request_failed — the
chokepoint every upstream failure path (OpenAI-compatible chat,
embeddings, responses, token counting, and Ollama) already routes
through — recording status, provider, url, model, error code, and the
upstream message truncated to 1000 chars. 4xx logs at WARNING, 5xx at
ERROR. Client-facing responses are unchanged, so no additional error
detail is exposed in the chat.


Claude-Session: https://claude.ai/code/session_018VecyiPejru1EVF5yfe2sU

Co-authored-by: Claude <noreply@anthropic.com>
---
 backend/open_webui/events.py | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/backend/open_webui/events.py b/backend/open_webui/events.py
index 942841ea3d..b627c69a18 100644
--- a/backend/open_webui/events.py
+++ b/backend/open_webui/events.py
@@ -1193,6 +1193,20 @@ async def publish_model_provider_request_failed(
         else 'upstream_error'
     )
 
+    # Server-log only; the upstream error body is otherwise invisible to admins
+    # (event sinks require an event function or webhook to be configured).
+    log.log(
+        logging.ERROR if status >= 500 else logging.WARNING,
+        'Upstream %s request failed: HTTP %d (%s) url=%s model=%s code=%s message=%s',
+        provider,
+        status,
+        error_type,
+        base_url,
+        requested_model or '-',
+        error_code or '-',
+        error_text[:MAX_STRING_LENGTH] or '-',
+    )
+
     data = {
         'error_type': error_type,
         'status': status,

From 48625e657ff11161c3588af2747598f102c1a4d1 Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 03:54:19 -0400
Subject: [PATCH 18/29] refac

---
 backend/open_webui/models/models.py           | 49 ++++++++++++++++++-
 .../workspace/Models/ModelEditor.svelte       | 23 ++++++++-
 2 files changed, 70 insertions(+), 2 deletions(-)

diff --git a/backend/open_webui/models/models.py b/backend/open_webui/models/models.py
index 9acd0c9b70..ff72dcfc65 100755
--- a/backend/open_webui/models/models.py
+++ b/backend/open_webui/models/models.py
@@ -3,7 +3,8 @@ from __future__ import annotations
 import json
 import logging
 import time
-from typing import Optional
+from copy import deepcopy
+from typing import Any, Optional
 
 from open_webui.internal.db import Base, JSONField, get_async_db_context
 from open_webui.models.access_grants import AccessGrantModel, AccessGrants
@@ -22,6 +23,38 @@ log = logging.getLogger(__name__)
 _warned_profile_urls: set[str] = set()
 
 
+def strip_extracted_content_from_model_knowledge(knowledge: Any) -> Any:
+    """Drop duplicated extracted text from ModelMeta.knowledge."""
+    if not isinstance(knowledge, list):
+        return knowledge
+
+    sanitized = []
+
+    for item in knowledge:
+        if not isinstance(item, dict):
+            sanitized.append(item)
+            continue
+
+        next_item = item
+        data = item.get('data')
+        if isinstance(data, dict) and 'content' in data:
+            next_item = deepcopy(item)
+            next_item.get('data', {}).pop('content', None)
+
+        file = next_item.get('file')
+        file_data = file.get('data') if isinstance(file, dict) else None
+        if isinstance(file_data, dict) and 'content' in file_data:
+            if next_item is item:
+                next_item = deepcopy(item)
+                file = next_item.get('file')
+                file_data = file.get('data') if isinstance(file, dict) else None
+            file_data.pop('content', None)
+
+        sanitized.append(next_item)
+
+    return sanitized
+
+
 # --- Models DB Schema ---
 
 
@@ -37,6 +70,7 @@ class ModelMeta(BaseModel):
     profile_image_url: str | None = None
     description: str | None = Field(default=None, description='User-facing description of the model.')
     capabilities: dict | None = None
+    knowledge: list[Any] | None = None
 
     model_config = ConfigDict(extra='allow')
 
@@ -56,6 +90,11 @@ class ModelMeta(BaseModel):
                 )
             return None
 
+    @field_validator('knowledge', mode='before')
+    @classmethod
+    def strip_knowledge_content(cls, v):
+        return strip_extracted_content_from_model_knowledge(v)
+
     @model_validator(mode='before')
     @classmethod
     def normalize_tags(cls, data):
@@ -152,6 +191,14 @@ class ModelsTable:
         access_grants: list[AccessGrantModel | None] = None,
         db: AsyncSession | None = None,
     ) -> ModelModel:
+        if isinstance(model.meta, dict):
+            knowledge = model.meta.get('knowledge')
+            stripped_knowledge = strip_extracted_content_from_model_knowledge(knowledge)
+            if stripped_knowledge != knowledge:
+                model.meta = {**model.meta, 'knowledge': stripped_knowledge}
+                if db is not None:
+                    await db.commit()
+
         model_data = ModelModel.model_validate(model).model_dump(exclude={'access_grants'})
         model_data['access_grants'] = (
             access_grants if access_grants is not None else await self._get_access_grants(model_data['id'], db=db)
diff --git a/src/lib/components/workspace/Models/ModelEditor.svelte b/src/lib/components/workspace/Models/ModelEditor.svelte
index f575096156..537abdb23b 100644
--- a/src/lib/components/workspace/Models/ModelEditor.svelte
+++ b/src/lib/components/workspace/Models/ModelEditor.svelte
@@ -148,6 +148,27 @@
 		voices = res?.voices ?? [];
 	};
 
+	const toModelKnowledgeReference = (item: any) => {
+		if (!item || typeof item !== 'object') {
+			return item;
+		}
+
+		return Object.fromEntries(
+			[
+				'id',
+				'name',
+				'type',
+				'description',
+				'context',
+				'legacy',
+				'collection_name',
+				'collection_names'
+			]
+				.filter((key) => item[key] !== undefined && item[key] !== null && item[key] !== '')
+				.map((key) => [key, item[key]])
+		);
+	};
+
 	const submitHandler = async () => {
 		loading = true;
 
@@ -194,7 +215,7 @@
 		}
 
 		if (knowledge.length > 0) {
-			info.meta.knowledge = knowledge;
+			info.meta.knowledge = knowledge.map(toModelKnowledgeReference);
 		} else {
 			if (info.meta.knowledge) {
 				delete info.meta.knowledge;

From cc9a44569ef08b64ff44d15607c43966f362ce75 Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 04:01:30 -0400
Subject: [PATCH 19/29] refac

---
 backend/open_webui/models/chats.py | 20 +++++++++++++++++---
 1 file changed, 17 insertions(+), 3 deletions(-)

diff --git a/backend/open_webui/models/chats.py b/backend/open_webui/models/chats.py
index 4f2213b84d..08aed47b22 100644
--- a/backend/open_webui/models/chats.py
+++ b/backend/open_webui/models/chats.py
@@ -1662,9 +1662,23 @@ class ChatTable:
                 postgres_content_sql = """
                 EXISTS (
                     SELECT 1
-                    FROM json_array_elements(Chat.chat->'messages') AS message
-                    WHERE json_typeof(message->'content') = 'string'
-                    AND LOWER(message->>'content') LIKE '%' || :content_key || '%'
+                    FROM chat_message AS message
+                    WHERE message.chat_id = Chat.id
+                    AND message.user_id = Chat.user_id
+                    AND json_typeof(message.content) = 'string'
+                    AND LOWER(message.content #>> '{}') LIKE '%' || :content_key || '%'
+                )
+                OR EXISTS (
+                    SELECT 1
+                    FROM json_each(Chat.chat#>'{history,messages}') AS history_message
+                    WHERE json_typeof(history_message.value->'content') = 'string'
+                    AND LOWER(history_message.value->>'content') LIKE '%' || :content_key || '%'
+                )
+                OR EXISTS (
+                    SELECT 1
+                    FROM json_array_elements(Chat.chat->'messages') AS legacy_message
+                    WHERE json_typeof(legacy_message->'content') = 'string'
+                    AND LOWER(legacy_message->>'content') LIKE '%' || :content_key || '%'
                 )
                 """
 

From bb12b1a18b77d80829cedb2d5bf965808222415b Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 04:16:14 -0400
Subject: [PATCH 20/29] refac

---
 backend/open_webui/main.py                   |  17 +-
 backend/open_webui/models/config.py          |   1 +
 backend/open_webui/utils/anthropic.py        | 154 +++++-
 src/lib/components/AddConnectionModal.svelte | 473 +++++++++++--------
 4 files changed, 434 insertions(+), 211 deletions(-)

diff --git a/backend/open_webui/main.py b/backend/open_webui/main.py
index 2f8542ec55..1b9ed9cc53 100644
--- a/backend/open_webui/main.py
+++ b/backend/open_webui/main.py
@@ -1795,8 +1795,23 @@ async def generate_messages(
         # Counting must not turn a compatible generation request into an outage.
         log.warning('Unable to count Anthropic input tokens for model %s', requested_model, exc_info=True)
 
+    model_id = requested_model
+    model_info = await Models.get_model_by_id(model_id)
+    if model_info and model_info.base_model_id:
+        model_id = model_info.base_model_id
+
+    passthrough_params = []
+    models = request.app.state.OPENAI_MODELS
+    if not models or model_id not in models:
+        await openai.get_all_models(request, user=user)
+        models = request.app.state.OPENAI_MODELS
+    model = models.get(model_id)
+    if model:
+        _, _, api_config = await openai.get_openai_connection(model['urlIdx'])
+        passthrough_params = api_config.get('passthrough_params') or []
+
     # Convert Anthropic payload to OpenAI format
-    openai_payload = convert_anthropic_to_openai_payload(form_data)
+    openai_payload = convert_anthropic_to_openai_payload(form_data, passthrough_params)
 
     # Route through the existing chat_completion handler
     response = await chat_completion(request, openai_payload, user)
diff --git a/backend/open_webui/models/config.py b/backend/open_webui/models/config.py
index 93a55f690d..5d0b2266b7 100644
--- a/backend/open_webui/models/config.py
+++ b/backend/open_webui/models/config.py
@@ -48,6 +48,7 @@ API_CONFIG_FIELDS = (
     'azure',
     'api_version',
     'extra_params',
+    'passthrough_params',
 )
 
 
diff --git a/backend/open_webui/utils/anthropic.py b/backend/open_webui/utils/anthropic.py
index 336767bd91..52dd2ea36a 100644
--- a/backend/open_webui/utils/anthropic.py
+++ b/backend/open_webui/utils/anthropic.py
@@ -12,6 +12,23 @@ from open_webui.utils.headers import include_user_info_headers
 
 log = logging.getLogger(__name__)
 
+ANTHROPIC_CONVERTED_REQUEST_PARAMS = {
+    'model',
+    'messages',
+    'system',
+    'max_tokens',
+    'temperature',
+    'top_p',
+    'top_k',
+    'stop_sequences',
+    'stream',
+    'metadata',
+    'service_tier',
+    'tools',
+    'tool_choice',
+    'reasoning_effort',
+}
+
 
 def is_anthropic_url(url: str) -> bool:
     """Check if the URL is an Anthropic API endpoint."""
@@ -109,7 +126,9 @@ def _finalize_openai_content(blocks: list) -> str | list:
     return blocks
 
 
-def convert_anthropic_to_openai_payload(anthropic_payload: dict) -> dict:
+def convert_anthropic_to_openai_payload(
+    anthropic_payload: dict, passthrough_params: list[str] | str | None = None
+) -> dict:
     """
     Convert an Anthropic Messages API request to OpenAI Chat Completions format.
 
@@ -350,6 +369,48 @@ def convert_anthropic_to_openai_payload(anthropic_payload: dict) -> dict:
     if 'max_tokens' in anthropic_payload:
         openai_payload['max_tokens'] = anthropic_payload['max_tokens']
 
+    captured_passthrough_params = {
+        param: value for param, value in anthropic_payload.items() if param not in ANTHROPIC_CONVERTED_REQUEST_PARAMS
+    }
+    if isinstance(passthrough_params, str):
+        passthrough_params = passthrough_params.split(',')
+    elif not isinstance(passthrough_params, (list, tuple, set)):
+        passthrough_params = []
+    passthrough_param_names = {str(item).strip() for item in passthrough_params if str(item).strip()}
+    if '*' in passthrough_param_names:
+        openai_payload.update(captured_passthrough_params)
+    else:
+        for param in passthrough_param_names:
+            if param in captured_passthrough_params:
+                openai_payload[param] = captured_passthrough_params[param]
+
+    output_config = anthropic_payload.get('output_config')
+    if isinstance(output_config, dict):
+        if 'effort' in output_config and 'reasoning_effort' not in anthropic_payload:
+            openai_payload['reasoning_effort'] = output_config['effort']
+
+        format_config = output_config.get('format')
+        if isinstance(format_config, dict):
+            format_type = format_config.get('type')
+            if format_type == 'json_schema':
+                json_schema = {
+                    'name': format_config.get('name', 'response_schema'),
+                    'schema': format_config.get('schema', {}),
+                }
+                if 'description' in format_config:
+                    json_schema['description'] = format_config['description']
+                if 'strict' in format_config:
+                    json_schema['strict'] = format_config['strict']
+                openai_payload['response_format'] = {
+                    'type': 'json_schema',
+                    'json_schema': json_schema,
+                }
+            elif format_type == 'json_object':
+                openai_payload['response_format'] = {'type': format_type}
+
+    if 'reasoning_effort' in anthropic_payload:
+        openai_payload['reasoning_effort'] = anthropic_payload['reasoning_effort']
+
     # Common parameters
     for param in ('temperature', 'top_p', 'top_k', 'stop_sequences', 'stream', 'metadata', 'service_tier'):
         if param in anthropic_payload:
@@ -421,6 +482,32 @@ def convert_openai_to_anthropic_response(
 
     # Build content blocks
     content = []
+    message_thinking = message.get('thinking')
+    thinking_blocks = message.get('thinking_blocks') or []
+    if not thinking_blocks and isinstance(message_thinking, dict):
+        thinking_blocks = message_thinking.get('blocks') or []
+
+    has_thinking = False
+    for block in thinking_blocks:
+        if not isinstance(block, dict):
+            continue
+
+        thinking = block.get('thinking') or block.get('content') or block.get('text')
+        if not thinking:
+            continue
+
+        thinking_block = {'type': 'thinking', 'thinking': thinking}
+        if block.get('signature'):
+            thinking_block['signature'] = block['signature']
+        content.append(thinking_block)
+        has_thinking = True
+
+    reasoning_content = message.get('reasoning_content') or message.get('reasoning')
+    if not reasoning_content and isinstance(message_thinking, str):
+        reasoning_content = message_thinking
+    if reasoning_content and not has_thinking:
+        content.append({'type': 'thinking', 'thinking': reasoning_content})
+
     message_content = message.get('content')
     if message_content:
         content.append({'type': 'text', 'text': message_content})
@@ -445,9 +532,11 @@ def convert_openai_to_anthropic_response(
     # Usage
     openai_usage = openai_response.get('usage', {})
     usage = {
-        'input_tokens': input_tokens
-        if input_tokens is not None
-        else openai_usage.get('input_tokens', openai_usage.get('prompt_tokens', 0)),
+        'input_tokens': (
+            input_tokens
+            if input_tokens is not None
+            else openai_usage.get('input_tokens', openai_usage.get('prompt_tokens', 0))
+        ),
         'output_tokens': openai_usage.get('output_tokens', openai_usage.get('completion_tokens', 0)),
     }
     if 'cache_creation_input_tokens' in openai_usage:
@@ -490,6 +579,7 @@ async def openai_stream_to_anthropic_stream(openai_stream_generator, model: str
     # Track content blocks with a running index.
     # Each text block or tool_use block gets its own index.
     current_block_index = 0
+    thinking_block_open = False
     text_block_open = False
 
     # Accumulated state for each tool call, keyed by tool call id.
@@ -563,11 +653,52 @@ async def openai_stream_to_anthropic_stream(openai_stream_generator, model: str
                         'output_tokens', data['usage'].get('completion_tokens', output_tokens)
                     )
 
+                reasoning_content = (
+                    delta.get('reasoning_content')
+                    or delta.get('reasoning')
+                    or delta.get('thinking')
+                    or message.get('reasoning_content')
+                    or message.get('reasoning')
+                )
+                if not reasoning_content:
+                    thinking_blocks = delta.get('thinking_blocks') or message.get('thinking_blocks') or []
+                    for block in thinking_blocks:
+                        if isinstance(block, dict):
+                            reasoning_content = block.get('thinking') or block.get('content') or block.get('text')
+                            if reasoning_content:
+                                break
+
+                if reasoning_content and not text_block_open and not has_tool_calls:
+                    if not thinking_block_open:
+                        block_start = {
+                            'type': 'content_block_start',
+                            'index': current_block_index,
+                            'content_block': {'type': 'thinking', 'thinking': ''},
+                        }
+                        yield f'event: content_block_start\ndata: {json.dumps(block_start)}\n\n'.encode()
+                        thinking_block_open = True
+
+                    block_delta = {
+                        'type': 'content_block_delta',
+                        'index': current_block_index,
+                        'delta': {'type': 'thinking_delta', 'thinking': reasoning_content},
+                    }
+                    yield f'event: content_block_delta\ndata: {json.dumps(block_delta)}\n\n'.encode()
+
                 # --- Handle text content ---
                 # Anthropic expects text blocks before tool blocks, so skip
                 # text deltas once any tool call has started.
                 content = delta.get('content')
                 if content and not has_tool_calls:
+                    if thinking_block_open:
+                        block_stop = {
+                            'type': 'content_block_stop',
+                            'index': current_block_index,
+                        }
+                        yield f'event: content_block_stop\ndata: {json.dumps(block_stop)}\n\n'.encode()
+                        thinking_block_open = False
+                        current_block_index += 1
+
                     if not text_block_open:
                         block_start = {
                             'type': 'content_block_start',
@@ -593,6 +724,15 @@ async def openai_stream_to_anthropic_stream(openai_stream_generator, model: str
 
                 if tool_calls:
                     # Close text block if one is open (text comes before tools)
+                    if thinking_block_open:
+                        block_stop = {
+                            'type': 'content_block_stop',
+                            'index': current_block_index,
+                        }
+                        yield f'event: content_block_stop\ndata: {json.dumps(block_stop)}\n\n'.encode()
+                        thinking_block_open = False
+                        current_block_index += 1
+
                     if text_block_open:
                         block_stop = {
                             'type': 'content_block_stop',
@@ -704,6 +844,12 @@ async def openai_stream_to_anthropic_stream(openai_stream_generator, model: str
     except Exception as e:
         log.error(f'Error in Anthropic stream conversion: {e}')
 
+    # Close any open thinking block
+    if thinking_block_open:
+        block_stop = {'type': 'content_block_stop', 'index': current_block_index}
+        yield f'event: content_block_stop\ndata: {json.dumps(block_stop)}\n\n'.encode()
+        current_block_index += 1
+
     # Flush any tools that buffered arguments but never emitted a block
     for tool in tracked_tool_calls.values():
         if not tool['started'] and tool['name']:
diff --git a/src/lib/components/AddConnectionModal.svelte b/src/lib/components/AddConnectionModal.svelte
index 8df0436283..6715c80b1a 100644
--- a/src/lib/components/AddConnectionModal.svelte
+++ b/src/lib/components/AddConnectionModal.svelte
@@ -49,6 +49,7 @@
 	let apiType = ''; // '' = chat completions (default), 'responses' = Responses API
 
 	let headers = '';
+	let passthroughParams = '';
 
 	let tags = [];
 
@@ -57,12 +58,19 @@
 
 	let loading = false;
 	let showDeleteConfirmDialog = false;
+	let showAdvanced = false;
 
 	const inputClass =
 		'bg-transparent outline-hidden placeholder:text-gray-300 dark:placeholder:text-gray-700';
 	const selectClass =
 		'dark:bg-gray-900 bg-transparent pr-5 outline-hidden placeholder:text-gray-300 dark:placeholder:text-gray-700';
 
+	const parsePassthroughParams = (value: string) =>
+		value
+			.split(',')
+			.map((param) => param.trim())
+			.filter(Boolean);
+
 	const verifyOllamaHandler = async () => {
 		// remove trailing slash from url
 		url = url.replace(/\/$/, '');
@@ -109,6 +117,7 @@
 					...(provider ? { provider } : {}),
 					...(azure ? { azure: true } : {}),
 					api_version: apiVersion,
+					passthrough_params: parsePassthroughParams(passthroughParams),
 					...(_headers ? { headers: _headers } : {})
 				}
 			},
@@ -149,6 +158,7 @@
 		if (azure) {
 			if (!apiVersion) {
 				loading = false;
+				showAdvanced = true;
 
 				toast.error($i18n.t('API Version is required'));
 				return;
@@ -163,6 +173,7 @@
 
 			if (modelIds.length === 0) {
 				loading = false;
+				showAdvanced = true;
 				toast.error($i18n.t('Deployment names are required for Azure OpenAI'));
 				return;
 			}
@@ -195,6 +206,7 @@
 				connection_type: connectionType,
 				auth_type,
 				headers: headers ? JSON.parse(headers) : undefined,
+				passthrough_params: parsePassthroughParams(passthroughParams),
 				...(provider ? { provider } : {}),
 				...(!ollama && azure ? { azure: true } : {}),
 				...(azure ? { api_version: apiVersion } : {}),
@@ -211,6 +223,8 @@
 		key = '';
 		auth_type = 'bearer';
 		prefixId = '';
+		passthroughParams = '';
+		showAdvanced = false;
 		tags = [];
 		modelIds = [];
 	};
@@ -228,6 +242,9 @@
 			enable = connection.config?.enable ?? true;
 			tags = connection.config?.tags ?? [];
 			prefixId = connection.config?.prefix_id ?? '';
+			passthroughParams = Array.isArray(connection.config?.passthrough_params)
+				? connection.config.passthrough_params.join(', ')
+				: (connection.config?.passthrough_params ?? '');
 			modelIds = connection.config?.model_ids ?? [];
 
 			if (ollama) {
@@ -429,107 +446,6 @@
 							</div>
 						</div>
 
-						{#if !direct}
-							<div class="flex gap-2 mt-2">
-								<div class="flex flex-col w-full">
-									<label
-										for="headers-input"
-										class={`mb-0.5 text-xs text-gray-500
-								`}>{$i18n.t('Headers')}</label
-									>
-
-									<div class="flex-1">
-										<Tooltip
-											content={$i18n.t(
-												'Enter additional headers in JSON format (e.g. {"X-Custom-Header": "value"}'
-											)}
-										>
-											<Textarea
-												className="w-full text-sm outline-hidden"
-												bind:value={headers}
-												placeholder={$i18n.t('Enter additional headers in JSON format')}
-												required={false}
-												minSize={30}
-											/>
-										</Tooltip>
-									</div>
-								</div>
-							</div>
-						{/if}
-
-						<div class="flex gap-2 mt-2">
-							<div class="flex flex-col w-full">
-								<label
-									for="prefix-id-input"
-									class={`mb-0.5 text-xs text-gray-500
-								`}>{$i18n.t('Prefix ID')}</label
-								>
-
-								<div class="flex-1">
-									<Tooltip
-										content={$i18n.t(
-											'Prefix ID is used to avoid conflicts with other connections by adding a prefix to the model IDs - leave empty to disable'
-										)}
-									>
-										<input
-											class={`w-full text-sm ${inputClass}`}
-											type="text"
-											id="prefix-id-input"
-											bind:value={prefixId}
-											placeholder={$i18n.t('Prefix ID')}
-											autocomplete="off"
-										/>
-									</Tooltip>
-								</div>
-							</div>
-						</div>
-
-						{#if !ollama && !direct}
-							<div class="flex flex-row justify-between items-center w-full mt-2">
-								<label
-									for="provider-select"
-									class={`mb-0.5 text-xs text-gray-500
-								`}>{$i18n.t('Provider')}</label
-								>
-
-								<div>
-									<select
-										id="provider-select"
-										bind:value={provider}
-										class="text-xs text-gray-700 dark:text-gray-300 bg-transparent outline-hidden"
-									>
-										<option value="">{$i18n.t('Default')}</option>
-										<option value="azure">{$i18n.t('Azure OpenAI')}</option>
-										<option value="llama.cpp">{$i18n.t('llama.cpp')}</option>
-									</select>
-								</div>
-							</div>
-						{/if}
-
-						{#if azure}
-							<div class="flex gap-2 mt-2">
-								<div class="flex flex-col w-full">
-									<label
-										for="api-version-input"
-										class={`mb-0.5 text-xs text-gray-500
-										`}>{$i18n.t('API Version')}</label
-									>
-
-									<div class="flex-1">
-										<input
-											id="api-version-input"
-											class={`w-full text-sm ${inputClass}`}
-											type="text"
-											bind:value={apiVersion}
-											placeholder={$i18n.t('API Version')}
-											autocomplete="off"
-											required
-										/>
-									</div>
-								</div>
-							</div>
-						{/if}
-
 						{#if !ollama && !direct}
 							<div class="flex flex-row justify-between items-center w-full mt-1">
 								<label
@@ -548,19 +464,7 @@
 										class=" text-xs text-gray-700 dark:text-gray-300"
 									>
 										{#if apiType === 'responses'}
-											<Tooltip
-												className="flex items-center gap-1"
-												content={$i18n.t(
-													'This feature is currently experimental and may not work as expected.'
-												)}
-											>
-												<span
-													class="inline-flex items-center text-[0.625rem] font-normal uppercase leading-none text-gray-400 dark:text-gray-600"
-													>{$i18n.t('Experimental')}</span
-												>
-
-												{$i18n.t('Responses')}
-											</Tooltip>
+											{$i18n.t('Responses')}
 										{:else}
 											{$i18n.t('Chat Completions')}
 										{/if}
@@ -569,111 +473,268 @@
 							</div>
 						{/if}
 
-						<div class="flex flex-col w-full mt-2">
-							<div class="mb-1 flex justify-between">
-								<div
-									class={`mb-0.5 text-xs text-gray-500
-								`}
+						<div class="flex items-center justify-between">
+							<button
+								type="button"
+								class="flex items-center gap-1 text-xs text-gray-500 dark:text-gray-400 hover:text-gray-700 dark:hover:text-gray-200 transition mt-2"
+								on:click={() => (showAdvanced = !showAdvanced)}
+							>
+								<svg
+									xmlns="http://www.w3.org/2000/svg"
+									viewBox="0 0 20 20"
+									fill="currentColor"
+									class="w-3 h-3 transition-transform {showAdvanced ? 'rotate-90' : ''}"
 								>
-									{$i18n.t('Model IDs')}
-								</div>
-							</div>
+									<path
+										fill-rule="evenodd"
+										d="M7.21 14.77a.75.75 0 01.02-1.06L11.168 10 7.23 6.29a.75.75 0 111.04-1.08l4.5 4.25a.75.75 0 010 1.08l-4.5 4.25a.75.75 0 01-1.06-.02z"
+										clip-rule="evenodd"
+									/>
+								</svg>
+								{$i18n.t('Advanced')}
+							</button>
+						</div>
 
-							{#if modelIds.length > 0}
-								<ul class="flex flex-col">
-									{#each modelIds as modelId, modelIdx}
-										<li class=" flex gap-2 w-full justify-between items-center">
-											<div class=" text-sm flex-1 py-1 rounded-lg">
-												{modelId}
-											</div>
-											<div class="shrink-0">
-												<button
-													aria-label={$i18n.t(`Remove {{MODELID}} from list.`, {
-														MODELID: modelId
-													})}
-													type="button"
-													on:click={() => {
-														modelIds = modelIds.filter((_, idx) => idx !== modelIdx);
-													}}
-												>
-													<Minus strokeWidth="2" className="size-3.5" />
-												</button>
-											</div>
-										</li>
-									{/each}
-								</ul>
-							{:else}
-								<div
-									class={`text-gray-500 text-xs text-center py-2 px-10
-								`}
-								>
-									{#if ollama}
-										{$i18n.t('Leave empty to include all models from "{{url}}/api/tags" endpoint', {
-											url: url
-										})}
-									{:else if azure}
-										{$i18n.t('Deployment names are required for Azure OpenAI')}
-										<!-- {$i18n.t('Leave empty to include all models from "{{url}}" endpoint', {
-											url: `${url}/openai/deployments`
-										})} -->
-									{:else}
-										{$i18n.t('Leave empty to include all models from "{{url}}/models" endpoint', {
-											url: url
-										})}
-									{/if}
+						{#if showAdvanced}
+							{#if !direct}
+								<div class="flex gap-2 mt-2">
+									<div class="flex flex-col w-full">
+										<label
+											for="headers-input"
+											class={`mb-0.5 text-xs text-gray-500
+								`}>{$i18n.t('Headers')}</label
+										>
+
+										<div class="flex-1">
+											<Tooltip
+												content={$i18n.t(
+													'Enter additional headers in JSON format (e.g. {"X-Custom-Header": "value"}'
+												)}
+											>
+												<Textarea
+													className="w-full text-sm outline-hidden"
+													bind:value={headers}
+													placeholder={$i18n.t('Enter additional headers in JSON format')}
+													required={false}
+													minSize={30}
+												/>
+											</Tooltip>
+										</div>
+									</div>
 								</div>
 							{/if}
-						</div>
 
-						<div class="flex items-center">
-							<label class="sr-only" for="add-model-id-input">{$i18n.t('Add a model ID')}</label>
-							<input
-								class={`w-full text-sm ${inputClass} ${modelId ? '' : 'text-gray-500'}`}
-								bind:value={modelId}
-								id="add-model-id-input"
-								placeholder={$i18n.t('Add a model ID')}
-							/>
+							{#if !ollama && !direct}
+								<div class="flex gap-2 mt-2">
+									<div class="flex flex-col w-full">
+										<label
+											for="allowed-passthrough-params-input"
+											class={`mb-0.5 text-xs text-gray-500
+								`}>{$i18n.t('Passthrough params')}</label
+										>
 
-							<div>
-								<button
-									type="button"
-									aria-label={$i18n.t('Add')}
-									on:click={() => {
-										addModelHandler();
-									}}
-								>
-									<Plus className="size-3.5" strokeWidth="2" />
-								</button>
+										<div class="flex-1">
+											<Tooltip
+												content={$i18n.t(
+													'Comma-separated top-level request parameters this upstream may receive without translation. Use * to allow all captured passthrough params.'
+												)}
+											>
+												<input
+													class={`w-full text-sm ${inputClass}`}
+													type="text"
+													id="allowed-passthrough-params-input"
+													bind:value={passthroughParams}
+													placeholder={$i18n.t('thinking, output_config')}
+													autocomplete="off"
+												/>
+											</Tooltip>
+										</div>
+									</div>
+								</div>
+							{/if}
+
+							<div class="flex gap-2 mt-2">
+								<div class="flex flex-col w-full">
+									<label
+										for="prefix-id-input"
+										class={`mb-0.5 text-xs text-gray-500
+								`}>{$i18n.t('Prefix ID')}</label
+									>
+
+									<div class="flex-1">
+										<Tooltip
+											content={$i18n.t(
+												'Prefix ID is used to avoid conflicts with other connections by adding a prefix to the model IDs - leave empty to disable'
+											)}
+										>
+											<input
+												class={`w-full text-sm ${inputClass}`}
+												type="text"
+												id="prefix-id-input"
+												bind:value={prefixId}
+												placeholder={$i18n.t('Prefix ID')}
+												autocomplete="off"
+											/>
+										</Tooltip>
+									</div>
+								</div>
 							</div>
-						</div>
-					</div>
 
-					<div class="flex gap-2 mt-2">
-						<div class="flex flex-col w-full">
-							<div
-								class={`mb-0.5 text-xs text-gray-500
+							{#if !ollama && !direct}
+								<div class="flex flex-row justify-between items-center w-full mt-2">
+									<label
+										for="provider-select"
+										class={`mb-0.5 text-xs text-gray-500
+								`}>{$i18n.t('Provider')}</label
+									>
+
+									<div>
+										<select
+											id="provider-select"
+											bind:value={provider}
+											class="text-xs text-gray-700 dark:text-gray-300 bg-transparent outline-hidden"
+										>
+											<option value="">{$i18n.t('Default')}</option>
+											<option value="azure">{$i18n.t('Azure OpenAI')}</option>
+											<option value="llama.cpp">{$i18n.t('llama.cpp')}</option>
+										</select>
+									</div>
+								</div>
+							{/if}
+
+							{#if azure}
+								<div class="flex gap-2 mt-2">
+									<div class="flex flex-col w-full">
+										<label
+											for="api-version-input"
+											class={`mb-0.5 text-xs text-gray-500
+										`}>{$i18n.t('API Version')}</label
+										>
+
+										<div class="flex-1">
+											<input
+												id="api-version-input"
+												class={`w-full text-sm ${inputClass}`}
+												type="text"
+												bind:value={apiVersion}
+												placeholder={$i18n.t('API Version')}
+												autocomplete="off"
+												required
+											/>
+										</div>
+									</div>
+								</div>
+							{/if}
+
+							<div class="flex flex-col w-full mt-2">
+								<div class="mb-1 flex justify-between">
+									<div
+										class={`mb-0.5 text-xs text-gray-500
 								`}
-							>
-								{$i18n.t('Tags')}
+									>
+										{$i18n.t('Model IDs')}
+									</div>
+								</div>
+
+								{#if modelIds.length > 0}
+									<ul class="flex flex-col">
+										{#each modelIds as modelId, modelIdx}
+											<li class=" flex gap-2 w-full justify-between items-center">
+												<div class=" text-sm flex-1 py-1 rounded-lg">
+													{modelId}
+												</div>
+												<div class="shrink-0">
+													<button
+														aria-label={$i18n.t(`Remove {{MODELID}} from list.`, {
+															MODELID: modelId
+														})}
+														type="button"
+														on:click={() => {
+															modelIds = modelIds.filter((_, idx) => idx !== modelIdx);
+														}}
+													>
+														<Minus strokeWidth="2" className="size-3.5" />
+													</button>
+												</div>
+											</li>
+										{/each}
+									</ul>
+								{:else}
+									<div
+										class={`text-gray-500 text-xs text-center py-2 px-10
+								`}
+									>
+										{#if ollama}
+											{$i18n.t(
+												'Leave empty to include all models from "{{url}}/api/tags" endpoint',
+												{
+													url: url
+												}
+											)}
+										{:else if azure}
+											{$i18n.t('Deployment names are required for Azure OpenAI')}
+											<!-- {$i18n.t('Leave empty to include all models from "{{url}}" endpoint', {
+											url: `${url}/openai/deployments`
+										})} -->
+										{:else}
+											{$i18n.t('Leave empty to include all models from "{{url}}/models" endpoint', {
+												url: url
+											})}
+										{/if}
+									</div>
+								{/if}
 							</div>
 
-							<div class="flex-1 mt-0.5">
-								<Tags
-									bind:tags
-									on:add={(e) => {
-										tags = [
-											...tags,
-											{
-												name: e.detail
-											}
-										];
-									}}
-									on:delete={(e) => {
-										tags = tags.filter((tag) => tag.name !== e.detail);
-									}}
+							<div class="flex items-center">
+								<label class="sr-only" for="add-model-id-input">{$i18n.t('Add a model ID')}</label>
+								<input
+									class={`w-full text-sm ${inputClass} ${modelId ? '' : 'text-gray-500'}`}
+									bind:value={modelId}
+									id="add-model-id-input"
+									placeholder={$i18n.t('Add a model ID')}
 								/>
+
+								<div>
+									<button
+										type="button"
+										aria-label={$i18n.t('Add')}
+										on:click={() => {
+											addModelHandler();
+										}}
+									>
+										<Plus className="size-3.5" strokeWidth="2" />
+									</button>
+								</div>
 							</div>
-						</div>
+
+							<div class="flex gap-2 mt-2">
+								<div class="flex flex-col w-full">
+									<div
+										class={`mb-0.5 text-xs text-gray-500
+								`}
+									>
+										{$i18n.t('Tags')}
+									</div>
+
+									<div class="flex-1 mt-0.5">
+										<Tags
+											bind:tags
+											on:add={(e) => {
+												tags = [
+													...tags,
+													{
+														name: e.detail
+													}
+												];
+											}}
+											on:delete={(e) => {
+												tags = tags.filter((tag) => tag.name !== e.detail);
+											}}
+										/>
+									</div>
+								</div>
+							</div>
+						{/if}
 					</div>
 
 					<div class="flex justify-between items-center pt-3 text-sm font-medium">

From d0f7da4f45b8831b09b2ab3ec8f91aa354d90ba3 Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 04:45:27 -0400
Subject: [PATCH 21/29] refac

---
 src/lib/components/chat/Messages.svelte       |   2 +
 .../chat/Messages/ContentRenderer.svelte      |   2 +-
 .../Markdown/ConsecutiveDetailsGroup.svelte   |   5 +-
 .../Messages/Markdown/MarkdownTokens.svelte   |  20 +-
 .../components/chat/Messages/Message.svelte   |   4 +
 .../Messages/MultiResponseMessages.svelte     |  30 ++-
 .../chat/Messages/ResponseMessage.svelte      |  36 +++-
 .../Messages/StructuredOutputRenderer.svelte  |  16 +-
 .../chat/Messages/UserMessage.svelte          |  14 +-
 .../layout/Sidebar/ChatHoverPreview.svelte    | 149 +++++++++++++++
 .../components/layout/Sidebar/ChatItem.svelte | 175 ++++++++++--------
 11 files changed, 343 insertions(+), 110 deletions(-)
 create mode 100644 src/lib/components/layout/Sidebar/ChatHoverPreview.svelte

diff --git a/src/lib/components/chat/Messages.svelte b/src/lib/components/chat/Messages.svelte
index 0adb4a027a..14d22d5b66 100644
--- a/src/lib/components/chat/Messages.svelte
+++ b/src/lib/components/chat/Messages.svelte
@@ -43,6 +43,7 @@
 	export let forkHandler: Function | null = null;
 
 	export let readOnly = false;
+	export let preview = false;
 	export let editCodeBlock = true;
 
 	export let topPadding = false;
@@ -543,6 +544,7 @@
 								{forkHandler}
 								{triggerScroll}
 								{readOnly}
+								{preview}
 								{editCodeBlock}
 								{topPadding}
 								{onInsertToNote}
diff --git a/src/lib/components/chat/Messages/ContentRenderer.svelte b/src/lib/components/chat/Messages/ContentRenderer.svelte
index dad616d39d..a6d3650cfb 100644
--- a/src/lib/components/chat/Messages/ContentRenderer.svelte
+++ b/src/lib/components/chat/Messages/ContentRenderer.svelte
@@ -307,7 +307,7 @@
 		{#if extracted.detailsContent}
 			<!-- Render structural blocks (tool calls, reasoning, etc.) through Markdown -->
 			<div class="markdown-prose">
-				<Markdown {id} content={extracted.detailsContent} {done} />
+				<Markdown {id} content={extracted.detailsContent} {preview} {done} />
 			</div>
 		{/if}
 		{#if extracted.plainContent}
diff --git a/src/lib/components/chat/Messages/Markdown/ConsecutiveDetailsGroup.svelte b/src/lib/components/chat/Messages/Markdown/ConsecutiveDetailsGroup.svelte
index fbc42f8b78..c9119a8687 100644
--- a/src/lib/components/chat/Messages/Markdown/ConsecutiveDetailsGroup.svelte
+++ b/src/lib/components/chat/Messages/Markdown/ConsecutiveDetailsGroup.svelte
@@ -31,6 +31,7 @@
 
 	export let messageDone = true;
 	export let allowEmbeds = true;
+	export let preview = false;
 
 	let open = $settings?.expandDetails ?? false;
 
@@ -112,7 +113,9 @@
 <div {id} class="w-full">
 	<!-- svelte-ignore a11y-no-static-element-interactions -->
 	<button
-		class="w-fit py-0.5 text-left text-[0.9375rem] text-gray-500 hover:text-gray-700 dark:hover:text-gray-300 transition cursor-pointer"
+		class="w-fit py-0.5 text-left {preview
+			? 'text-xs'
+			: 'text-[0.9375rem]'} text-gray-500 hover:text-gray-700 dark:hover:text-gray-300 transition cursor-pointer"
 		aria-label={$i18n.t('Toggle details')}
 		aria-expanded={open}
 		on:click={() => {
diff --git a/src/lib/components/chat/Messages/Markdown/MarkdownTokens.svelte b/src/lib/components/chat/Messages/Markdown/MarkdownTokens.svelte
index ac487cec25..9f391c3e7c 100644
--- a/src/lib/components/chat/Messages/Markdown/MarkdownTokens.svelte
+++ b/src/lib/components/chat/Messages/Markdown/MarkdownTokens.svelte
@@ -97,6 +97,10 @@
 			.trim();
 	};
 
+	$: detailButtonClassName = `w-fit py-0.5 ${
+		preview ? 'text-xs' : 'text-[0.9375rem]'
+	} text-gray-500 hover:text-gray-700 dark:hover:text-gray-300 transition`;
+
 	$: displayTokens = getDisplayTokens(tokens);
 	$: singlePlainBlock =
 		displayTokens.length === 1 &&
@@ -278,6 +282,7 @@
 					id={`${id}-${tokenIdx}`}
 					tokens={token.tokens}
 					{done}
+					{preview}
 					{editCodeBlock}
 					{onTaskClick}
 					{sourceIds}
@@ -313,6 +318,7 @@
 							tokens={item.tokens}
 							top={token.loose}
 							{done}
+							{preview}
 							{editCodeBlock}
 							{onTaskClick}
 							{sourceIds}
@@ -348,6 +354,7 @@
 									tokens={item.tokens}
 									top={token.loose}
 									{done}
+									{preview}
 									{editCodeBlock}
 									{onTaskClick}
 									{sourceIds}
@@ -360,6 +367,7 @@
 								tokens={item.tokens}
 								top={token.loose}
 								{done}
+								{preview}
 								{editCodeBlock}
 								{onTaskClick}
 								{sourceIds}
@@ -375,6 +383,7 @@
 			id={`${id}-${tokenIdx}-detail-group`}
 			tokens={token.items}
 			messageDone={done}
+			{preview}
 			{allowEmbeds}
 		>
 			<div slot="content">
@@ -389,7 +398,7 @@
 							grouped={true}
 							open={$settings?.expandDetails ?? false}
 							className="w-full"
-							buttonClassName="w-fit py-0.5 text-[0.9375rem] text-gray-500 hover:text-gray-700 dark:hover:text-gray-300 transition"
+							buttonClassName={detailButtonClassName}
 						/>
 					{:else if textContent.length > 0}
 						<Collapsible
@@ -398,7 +407,7 @@
 							attributes={detailToken?.attributes}
 							messageDone={done}
 							className="w-full"
-							buttonClassName="w-fit py-0.5 text-[0.9375rem] text-gray-500 hover:text-gray-700 dark:hover:text-gray-300 transition"
+							buttonClassName={detailButtonClassName}
 							dir="auto"
 						>
 							<div class="mb-1.5" slot="content">
@@ -407,6 +416,7 @@
 									tokens={marked.lexer(decode(detailToken.text))}
 									attributes={detailToken?.attributes}
 									{done}
+									{preview}
 									{editCodeBlock}
 									{onTaskClick}
 									{sourceIds}
@@ -422,7 +432,7 @@
 							attributes={detailToken?.attributes}
 							messageDone={done}
 							className="w-full"
-							buttonClassName="w-fit py-0.5 text-[0.9375rem] text-gray-500 hover:text-gray-700 dark:hover:text-gray-300 transition"
+							buttonClassName={detailButtonClassName}
 							dir="auto"
 						/>
 					{/if}
@@ -440,6 +450,7 @@
 				resultContent={getDetailTextContent(token)}
 				open={$settings?.expandDetails ?? false}
 				className="w-full space-y-2"
+				buttonClassName={detailButtonClassName}
 			/>
 		{:else if textContent.length > 0}
 			<Collapsible
@@ -448,6 +459,7 @@
 				attributes={token?.attributes}
 				messageDone={done}
 				className="w-full space-y-2"
+				buttonClassName={detailButtonClassName}
 				dir="auto"
 			>
 				<div class=" mb-1.5" slot="content">
@@ -456,6 +468,7 @@
 						tokens={marked.lexer(decode(token.text))}
 						attributes={token?.attributes}
 						{done}
+						{preview}
 						{editCodeBlock}
 						{onTaskClick}
 						{sourceIds}
@@ -471,6 +484,7 @@
 				attributes={token?.attributes}
 				messageDone={done}
 				className="w-full space-y-2"
+				buttonClassName={detailButtonClassName}
 				dir="auto"
 			/>
 		{/if}
diff --git a/src/lib/components/chat/Messages/Message.svelte b/src/lib/components/chat/Messages/Message.svelte
index 0b18ca2d9f..f7839bea8c 100644
--- a/src/lib/components/chat/Messages/Message.svelte
+++ b/src/lib/components/chat/Messages/Message.svelte
@@ -42,6 +42,7 @@
 	export let forkHandler: Function | null = null;
 	export let triggerScroll;
 	export let readOnly = false;
+	export let preview = false;
 	export let editCodeBlock = true;
 	export let topPadding = false;
 	export let onInsertToNote: ((content: string) => void) | null = null;
@@ -72,6 +73,7 @@
 				{editMessage}
 				{deleteMessage}
 				{readOnly}
+				{preview}
 				{editCodeBlock}
 				{topPadding}
 				{onInsertToNote}
@@ -100,6 +102,7 @@
 				{addMessages}
 				{forkHandler}
 				{readOnly}
+				{preview}
 				{editCodeBlock}
 				{topPadding}
 			/>
@@ -126,6 +129,7 @@
 					{addMessages}
 					{forkHandler}
 					{readOnly}
+					{preview}
 					{editCodeBlock}
 					{topPadding}
 					{onInsertToNote}
diff --git a/src/lib/components/chat/Messages/MultiResponseMessages.svelte b/src/lib/components/chat/Messages/MultiResponseMessages.svelte
index 6223cf8338..a2ce3f0f2f 100644
--- a/src/lib/components/chat/Messages/MultiResponseMessages.svelte
+++ b/src/lib/components/chat/Messages/MultiResponseMessages.svelte
@@ -28,6 +28,7 @@
 
 	export let isLastMessage;
 	export let readOnly = false;
+	export let preview = false;
 	export let editCodeBlock = true;
 
 	export let setInputText: Function = () => {};
@@ -251,7 +252,11 @@
 		>
 			{#if $settings?.displayMultiModelResponsesInTabs ?? false}
 				<div class="w-full">
-					<div class=" flex w-full mb-4.5 border-b border-gray-200 dark:border-gray-850">
+					<div
+						class=" flex w-full mb-4.5 border-b border-gray-200 dark:border-gray-850 {preview
+							? 'hidden'
+							: ''}"
+					>
 						<div
 							class="flex gap-2 scrollbar-none overflow-x-auto w-fit text-center font-normal bg-transparent pt-1 text-sm"
 							on:wheel|preventDefault={(e) => {
@@ -322,6 +327,7 @@
 									{addMessages}
 									{forkHandler}
 									{readOnly}
+									{preview}
 									{topPadding}
 									{onInsertToNote}
 								/>
@@ -338,14 +344,17 @@
 							groupedMessageIds[modelIdx].messageIds[groupedMessageIdsIdx[modelIdx]]}
 
 						<div
-							class=" snap-center w-full max-w-full m-1 border {history.messages[messageId]
-								?.modelIdx == modelIdx
-								? `bg-gray-50 dark:bg-gray-850 border-gray-100 dark:border-gray-800 border-2 ${
-										$mobile ? 'min-w-full' : 'min-w-80'
-									}`
-								: `border-gray-100/30 dark:border-gray-850/30 border-dashed ${
-										$mobile ? 'min-w-full' : 'min-w-80'
-									}`} transition-all p-5 rounded-2xl"
+							class="snap-center w-full max-w-full transition-all {preview
+								? ''
+								: `m-1 border p-5 rounded-2xl ${
+										history.messages[messageId]?.modelIdx == modelIdx
+											? `bg-gray-50 dark:bg-gray-850 border-gray-100 dark:border-gray-800 border-2 ${
+													$mobile ? 'min-w-full' : 'min-w-80'
+												}`
+											: `border-gray-100/30 dark:border-gray-850/30 border-dashed ${
+													$mobile ? 'min-w-full' : 'min-w-80'
+												}`
+									}`}"
 							on:click={async () => {
 								onGroupClick(_messageId, modelIdx);
 							}}
@@ -380,6 +389,7 @@
 										{addMessages}
 										{forkHandler}
 										{readOnly}
+										{preview}
 										{editCodeBlock}
 										{topPadding}
 										{onInsertToNote}
@@ -392,7 +402,7 @@
 			{/if}
 		</div>
 
-		{#if !readOnly}
+		{#if !preview && !readOnly}
 			{#if !Object.keys(groupedMessageIds).find((modelIdx) => {
 				const { messageIds } = groupedMessageIds[modelIdx];
 				const _messageId = messageIds[groupedMessageIdsIdx[modelIdx]];
diff --git a/src/lib/components/chat/Messages/ResponseMessage.svelte b/src/lib/components/chat/Messages/ResponseMessage.svelte
index 36d05bc37b..33bf324f81 100644
--- a/src/lib/components/chat/Messages/ResponseMessage.svelte
+++ b/src/lib/components/chat/Messages/ResponseMessage.svelte
@@ -165,6 +165,7 @@
 
 	export let isLastMessage = true;
 	export let readOnly = false;
+	export let preview = false;
 	export let editCodeBlock = true;
 	export let topPadding = false;
 	export let onInsertToNote: ((content: string) => void) | null = null;
@@ -665,13 +666,15 @@
 		</div>
 
 		<div class="flex-auto w-0 pl-1 relative">
-			<Name>
-				<Tooltip content={model?.name ?? message.model} placement="top-start">
-					<span id="response-message-model-name" class="line-clamp-1 text-black dark:text-white">
-						{model?.name ?? message.model}
-					</span>
-				</Tooltip>
-			</Name>
+			{#if !preview}
+				<Name>
+					<Tooltip content={model?.name ?? message.model} placement="top-start">
+						<span id="response-message-model-name" class="line-clamp-1 text-black dark:text-white">
+							{model?.name ?? message.model}
+						</span>
+					</Tooltip>
+				</Name>
+			{/if}
 
 			<div>
 				<div class="chat-{message.role} w-full min-w-full">
@@ -822,7 +825,7 @@
 										!readOnly &&
 										($settings?.showFloatingActionButtons ?? true)}
 									save={!readOnly}
-									preview={!readOnly}
+									preview={preview || !readOnly}
 									{editCodeBlock}
 									{topPadding}
 									done={($settings?.chatFadeStreamingText ?? true)
@@ -891,7 +894,22 @@
 					</div>
 				</div>
 
-				{#if !edit}
+				{#if preview && message.timestamp}
+					<div class="mt-0.5 flex justify-start whitespace-nowrap text-gray-600 dark:text-gray-500">
+						<Tooltip
+							className="flex self-center"
+							content={formatMessageTimestampFull(message.timestamp * 1000)}
+							placement="bottom"
+						>
+							<time
+								datetime={new Date(message.timestamp * 1000).toISOString()}
+								class="ml-1 shrink-0 whitespace-nowrap text-[0.6875rem] tabular-nums text-gray-400 dark:text-gray-600 select-none"
+							>
+								{formatMessageTimestamp(message.timestamp * 1000)}
+							</time>
+						</Tooltip>
+					</div>
+				{:else if !edit}
 					<div
 						bind:this={buttonsContainerElement}
 						class="flex items-center justify-start overflow-x-auto whitespace-nowrap buttons text-gray-600 dark:text-gray-500 mt-0.5 [&>*]:shrink-0"
diff --git a/src/lib/components/chat/Messages/StructuredOutputRenderer.svelte b/src/lib/components/chat/Messages/StructuredOutputRenderer.svelte
index ae88360075..a39e8dcbdb 100644
--- a/src/lib/components/chat/Messages/StructuredOutputRenderer.svelte
+++ b/src/lib/components/chat/Messages/StructuredOutputRenderer.svelte
@@ -32,6 +32,10 @@
 	const getDetailTitle = (detailToken: OutputDetailToken): any => detailToken.summary;
 	const getDetailAttributes = (detailToken: OutputDetailToken): any => detailToken.attributes;
 
+	$: detailButtonClassName = `w-fit py-0.5 ${
+		preview ? 'text-xs' : 'text-[0.9375rem]'
+	} text-gray-500 hover:text-gray-700 dark:hover:text-gray-300 transition`;
+
 	$: displayItems = buildOutputDisplayItems(output) as OutputDisplayItem[];
 </script>
 
@@ -64,6 +68,7 @@
 			id={`${id}-${displayItem.id}`}
 			tokens={displayItem.tokens}
 			messageDone={done}
+			{preview}
 		>
 			<div slot="content">
 				{#each displayItem.tokens as detailToken, detailIndex}
@@ -75,7 +80,7 @@
 							grouped={true}
 							open={$settings?.expandDetails ?? false}
 							className="w-full"
-							buttonClassName="w-fit py-0.5 text-[0.9375rem] text-gray-500 hover:text-gray-700 dark:hover:text-gray-300 transition"
+							buttonClassName={detailButtonClassName}
 						/>
 					{:else if detailToken.text?.length > 0}
 						<Collapsible
@@ -84,7 +89,7 @@
 							attributes={getDetailAttributes(detailToken)}
 							messageDone={done}
 							className="w-full"
-							buttonClassName="w-fit py-0.5 text-[0.9375rem] text-gray-500 hover:text-gray-700 dark:hover:text-gray-300 transition"
+							buttonClassName={detailButtonClassName}
 						>
 							<div class="mb-1.5" slot="content">
 								<div class="markdown-prose">
@@ -92,6 +97,7 @@
 										id={`${id}-${displayItem.id}-${detailIndex}-detail`}
 										content={detailToken.text}
 										{done}
+										{preview}
 										{editCodeBlock}
 									/>
 								</div>
@@ -105,7 +111,7 @@
 							attributes={getDetailAttributes(detailToken)}
 							messageDone={done}
 							className="w-full"
-							buttonClassName="w-fit py-0.5 text-[0.9375rem] text-gray-500 hover:text-gray-700 dark:hover:text-gray-300 transition"
+							buttonClassName={detailButtonClassName}
 						/>
 					{/if}
 				{/each}
@@ -120,6 +126,7 @@
 				resultContent={detailToken.text}
 				open={$settings?.expandDetails ?? false}
 				className="w-full space-y-2"
+				buttonClassName={detailButtonClassName}
 			/>
 		{:else if detailToken.text?.length > 0}
 			<Collapsible
@@ -128,6 +135,7 @@
 				attributes={getDetailAttributes(detailToken)}
 				messageDone={done}
 				className="w-full space-y-2"
+				buttonClassName={detailButtonClassName}
 			>
 				<div class="mb-1.5" slot="content">
 					<div class="markdown-prose">
@@ -135,6 +143,7 @@
 							id={`${id}-${displayItem.id}-detail`}
 							content={detailToken.text}
 							{done}
+							{preview}
 							{editCodeBlock}
 						/>
 					</div>
@@ -148,6 +157,7 @@
 				attributes={getDetailAttributes(detailToken)}
 				messageDone={done}
 				className="w-full space-y-2"
+				buttonClassName={detailButtonClassName}
 			/>
 		{/if}
 	{/if}
diff --git a/src/lib/components/chat/Messages/UserMessage.svelte b/src/lib/components/chat/Messages/UserMessage.svelte
index 5c4ead8ff5..21cb2f44b2 100644
--- a/src/lib/components/chat/Messages/UserMessage.svelte
+++ b/src/lib/components/chat/Messages/UserMessage.svelte
@@ -39,6 +39,7 @@
 
 	export let isFirstMessage: boolean;
 	export let readOnly: boolean;
+	export let preview = false;
 	export let editCodeBlock = true;
 	export let topPadding = false;
 
@@ -415,7 +416,8 @@
 						>
 							<time
 								datetime={new Date(message.timestamp * 1000).toISOString()}
-								class="invisible group-hover:visible {($settings?.chatBubble ?? true)
+								class="{preview ? '' : 'invisible group-hover:visible'} {($settings?.chatBubble ??
+								true)
 									? 'mr-1'
 									: 'ml-1 shrink-0 whitespace-nowrap'} text-[0.6875rem] tabular-nums text-gray-400 dark:text-gray-600 select-none"
 							>
@@ -424,7 +426,7 @@
 						</Tooltip>
 					{/if}
 
-					{#if !($settings?.chatBubble ?? true)}
+					{#if !preview && !($settings?.chatBubble ?? true)}
 						{#if siblings.length > 1}
 							<div class="flex self-center" dir="ltr">
 								<button
@@ -518,7 +520,7 @@
 							</div>
 						{/if}
 					{/if}
-					{#if !readOnly}
+					{#if !preview && !readOnly}
 						<Tooltip content={$i18n.t('Edit')} placement="bottom">
 							<button
 								class="{($settings?.highContrastMode ?? false)
@@ -546,7 +548,7 @@
 						</Tooltip>
 					{/if}
 
-					{#if message?.content}
+					{#if !preview && message?.content}
 						<Tooltip content={$i18n.t('Copy')} placement="bottom">
 							<button
 								class="{($settings?.highContrastMode ?? false)
@@ -575,7 +577,7 @@
 					{/if}
 
 					{#if $_user?.role === 'admin' || ($_user?.permissions?.chat?.delete_message ?? false)}
-						{#if !readOnly && (!isFirstMessage || siblings.length > 1)}
+						{#if !preview && !readOnly && (!isFirstMessage || siblings.length > 1)}
 							<Tooltip content={$i18n.t('Delete')} placement="bottom">
 								<button
 									class="{($settings?.highContrastMode ?? false)
@@ -608,7 +610,7 @@
 						{/if}
 					{/if}
 
-					{#if $settings?.chatBubble ?? true}
+					{#if !preview && ($settings?.chatBubble ?? true)}
 						{#if siblings.length > 1}
 							<div class="flex self-center" dir="ltr">
 								<button
diff --git a/src/lib/components/layout/Sidebar/ChatHoverPreview.svelte b/src/lib/components/layout/Sidebar/ChatHoverPreview.svelte
new file mode 100644
index 0000000000..1a45373a45
--- /dev/null
+++ b/src/lib/components/layout/Sidebar/ChatHoverPreview.svelte
@@ -0,0 +1,149 @@
+<script lang="ts">
+	import { getContext, onDestroy, tick } from 'svelte';
+	import { LinkPreview } from 'bits-ui';
+
+	import { getChatById } from '$lib/apis/chats';
+	import Messages from '$lib/components/chat/Messages.svelte';
+	import Spinner from '$lib/components/common/Spinner.svelte';
+	import { user } from '$lib/stores';
+	import { createMessagesList } from '$lib/utils';
+
+	const i18n: any = getContext('i18n');
+
+	export let chatId = '';
+	export let title = '';
+	export let openPreview = false;
+	export let side: 'left' | 'right' | 'bottom' | 'top' = 'right';
+	export let align: 'start' | 'end' | 'center' = 'start';
+	export let sideOffset = 12;
+
+	let messagesContainerElement: HTMLElement | null = null;
+
+	let selectedModels = [''];
+	let history: any = null;
+	let previewMessages: any[] = [];
+	let previewReady = false;
+	let loading = false;
+	let error = '';
+	let requestedChatId = '';
+	let loadToken = 0;
+
+	const scrollPreviewToBottom = async () => {
+		await tick();
+		requestAnimationFrame(() => {
+			if (messagesContainerElement) {
+				messagesContainerElement.scrollTop = messagesContainerElement.scrollHeight;
+
+				requestAnimationFrame(() => {
+					if (messagesContainerElement) {
+						messagesContainerElement.scrollTop = messagesContainerElement.scrollHeight;
+					}
+				});
+			}
+		});
+		setTimeout(() => {
+			if (messagesContainerElement) {
+				messagesContainerElement.scrollTop = messagesContainerElement.scrollHeight;
+			}
+		}, 80);
+	};
+
+	const loadChatPreview = async (id: string) => {
+		if (!id || loading || requestedChatId === id) return;
+
+		const token = ++loadToken;
+		requestedChatId = id;
+		loading = true;
+		error = '';
+		previewReady = false;
+		previewMessages = [];
+		history = null;
+		selectedModels = [''];
+
+		const chat = await getChatById(localStorage.token, id).catch(() => null);
+		if (token !== loadToken) return;
+
+		if (chat?.chat?.history) {
+			selectedModels = Array.isArray(chat.chat.models)
+				? chat.chat.models
+				: [chat.chat.models ?? ''];
+			history = chat.chat.history;
+			previewMessages = createMessagesList(chat.chat.history, chat.chat.history.currentId);
+			previewReady = true;
+			scrollPreviewToBottom();
+		} else if (chat) {
+			previewMessages = [];
+			previewReady = true;
+		} else {
+			error = $i18n.t('Failed to load chat preview');
+		}
+
+		loading = false;
+	};
+
+	$: if (openPreview && chatId && requestedChatId !== chatId) {
+		loadChatPreview(chatId);
+	}
+
+	$: if (openPreview && previewReady) {
+		scrollPreviewToBottom();
+	}
+
+	onDestroy(() => {
+		loadToken += 1;
+	});
+</script>
+
+{#if openPreview}
+	<LinkPreview.Portal>
+		<LinkPreview.Content
+			class="z-[9999] hidden max-h-[min(17.5rem,calc(100vh-1.5rem))] w-[20rem] max-w-[calc(100vw-1.5rem)] overflow-hidden rounded-2xl bg-white shadow-[0_16px_40px_-28px_rgba(0,0,0,0.55)] ring-1 ring-black/5 transition md:block dark:bg-gray-850 dark:text-white dark:ring-white/10"
+			{side}
+			{align}
+			{sideOffset}
+		>
+			<div class="border-b border-gray-50/60 px-3 py-2 dark:border-gray-800/25">
+				<div class="truncate text-[13px] font-medium leading-5 text-gray-900 dark:text-gray-100">
+					{title || $i18n.t('Chat')}
+				</div>
+			</div>
+			<div
+				bind:this={messagesContainerElement}
+				class="max-h-[min(15rem,calc(100vh-4rem))] overflow-y-auto bg-white scrollbar-hover @container dark:bg-gray-850"
+			>
+				{#if loading}
+					<div
+						class="flex w-full items-center justify-center py-8 text-gray-500 dark:text-gray-400"
+					>
+						<Spinner className="size-5" />
+					</div>
+				{:else if error}
+					<div
+						class="flex w-full items-center justify-center px-6 py-6 text-center text-sm text-gray-500 dark:text-gray-400"
+					>
+						{error}
+					</div>
+				{:else if previewReady}
+					<Messages
+						className="flex w-full pt-2 pb-0 [&_.message-listitem]:!mb-1 [&_.message-listitem]:!max-w-none [&_.message-listitem]:!px-3 [&_.pb-18]:!pb-1.5 [&_.markdown-prose]:!text-xs [&_.markdown-prose]:!leading-snug [&_.whitespace-pre-wrap]:!text-xs [&_.whitespace-pre-wrap]:!leading-snug [&_.text-\[0\.9375rem\]]:!text-xs [&_.text-sm]:!text-xs [&_.tool-call-body_pre]:!text-[11px] [&_.rounded-3xl]:!rounded-2xl [&_.chat-user_.rounded-3xl]:!bg-gray-50 dark:[&_.chat-user_.rounded-3xl]:!bg-gray-800 [&_.px-4]:!px-3 [&_.py-3]:!py-2 [&_.py-1\.5]:!py-1"
+						chatId={`chat-hover-preview-${chatId}`}
+						user={$user}
+						prompt=""
+						readOnly={true}
+						preview={true}
+						{selectedModels}
+						atSelectedModel={null}
+						{history}
+						autoScroll={true}
+						messagesCount={null}
+						sendMessage={() => {}}
+						continueResponse={() => {}}
+						regenerateResponse={() => {}}
+						mergeResponses={() => {}}
+						chatActionHandler={() => {}}
+					/>
+				{/if}
+			</div>
+		</LinkPreview.Content>
+	</LinkPreview.Portal>
+{/if}
diff --git a/src/lib/components/layout/Sidebar/ChatItem.svelte b/src/lib/components/layout/Sidebar/ChatItem.svelte
index acb2a5bae7..65569af1c6 100644
--- a/src/lib/components/layout/Sidebar/ChatItem.svelte
+++ b/src/lib/components/layout/Sidebar/ChatItem.svelte
@@ -9,6 +9,7 @@
 	import { toast } from 'svelte-sonner';
 	import { goto, invalidate, invalidateAll } from '$app/navigation';
 	import { onMount, getContext, createEventDispatcher, tick } from 'svelte';
+	import { LinkPreview } from 'bits-ui';
 	import {
 		archiveChatById,
 		cloneChatById,
@@ -37,6 +38,7 @@
 	import Tooltip from '$lib/components/common/Tooltip.svelte';
 	import DragGhost from '$lib/components/common/DragGhost.svelte';
 	import Spinner from '$lib/components/common/Spinner.svelte';
+	import ChatHoverPreview from './ChatHoverPreview.svelte';
 	import ChatIcon from './icons/Chat.svelte';
 	import MoreHorizontalIcon from './icons/MoreHorizontal.svelte';
 	import SparklesIcon from './icons/Sparkles.svelte';
@@ -90,6 +92,7 @@
 	let chat = null;
 
 	let mouseOver = false;
+	let openPreview = false;
 
 	// Local state: tracks the last updatedAt seen while the user was viewing
 	// this chat.  Survives prop refreshes from sidebar data re-fetches that
@@ -247,6 +250,7 @@
 
 	const onDragStart = (event) => {
 		event.stopPropagation();
+		openPreview = false;
 
 		event.dataTransfer.setDragImage(invisibleDragImage, 0, 0);
 
@@ -329,6 +333,7 @@
 	const renameHandler = async () => {
 		chatTitle = title;
 		confirmEdit = true;
+		openPreview = false;
 
 		await tick();
 
@@ -444,11 +449,18 @@
 	</DragGhost>
 {/if}
 
+<!-- svelte-ignore a11y-no-static-element-interactions -->
 <div
 	id="sidebar-chat-group"
 	bind:this={itemElement}
 	class=" w-full {className} relative group"
 	draggable={!confirmEdit && !readonly}
+	on:mouseenter={() => {
+		mouseOver = true;
+	}}
+	on:mouseleave={() => {
+		mouseOver = false;
+	}}
 >
 	{#if confirmEdit}
 		<div
@@ -490,90 +502,99 @@
 			/>
 		</div>
 	{:else}
-		<a
-			id="sidebar-chat-item"
-			class=" w-full flex justify-between rounded-xl px-2 py-[6px] {id === $chatId || confirmEdit
-				? ($settings?.highContrastMode ?? false)
-					? 'bg-black/[0.035] dark:bg-white/[0.06] selected'
-					: 'bg-black/[0.035] dark:bg-white/[0.045] selected'
-				: selected
-					? ($settings?.highContrastMode ?? false)
-						? 'bg-black/[0.035] dark:bg-white/[0.055] selected'
-						: 'bg-black/[0.035] dark:bg-white/[0.045] selected'
-					: ' hover:bg-gray-50 dark:hover:bg-gray-900 group-hover:bg-gray-50 dark:group-hover:bg-gray-900'}  whitespace-nowrap text-ellipsis transition"
-			href="/c/{id}"
-			on:click={() => {
-				dispatch('select');
-
-				if ($selectedFolder) {
-					selectedFolder.set(null);
-				}
-
-				if ($mobile) {
-					showSidebar.set(false);
-				}
-
-				// Optimistically mark as read in UI when clicked
-				unread = false;
-				lastReadAt = Date.now() / 1000;
-			}}
-			on:dblclick={async (e) => {
-				if (readonly) return;
-				e.preventDefault();
-				e.stopPropagation();
-
-				doubleClicked = true;
-				renameHandler();
-			}}
-			on:mouseenter={(e) => {
-				mouseOver = true;
-			}}
-			on:mouseleave={(e) => {
-				mouseOver = false;
-			}}
-			on:focus={(e) => {}}
-			draggable="false"
+		<LinkPreview.Root
+			openDelay={300}
+			closeDelay={0}
+			disabled={$mobile || confirmEdit || dragged}
+			bind:open={openPreview}
 		>
-			{#if ownerUserId}
-				<Tooltip content={ownerName || 'Unknown'}>
-					<img
-						src="/api/v1/users/{ownerUserId}/profile/image"
-						alt=""
-						class="size-3.5 rounded-full shrink-0 object-cover mr-1.5"
-					/>
-				</Tooltip>
-			{/if}
+			<LinkPreview.Trigger
+				id="sidebar-chat-item"
+				class=" w-full flex justify-between rounded-xl px-2 py-[6px] {id === $chatId || confirmEdit
+					? ($settings?.highContrastMode ?? false)
+						? 'bg-black/[0.035] dark:bg-white/[0.06] selected'
+						: 'bg-black/[0.035] dark:bg-white/[0.045] selected'
+					: selected
+						? ($settings?.highContrastMode ?? false)
+							? 'bg-black/[0.035] dark:bg-white/[0.055] selected'
+							: 'bg-black/[0.035] dark:bg-white/[0.045] selected'
+						: ' hover:bg-gray-50 dark:hover:bg-gray-900 group-hover:bg-gray-50 dark:group-hover:bg-gray-900'}  whitespace-nowrap text-ellipsis transition"
+				href="/c/{id}"
+				onclick={() => {
+					openPreview = false;
+					dispatch('select');
 
-			<!-- Loading spinner for active chat (left side) -->
-			{#if active}
-				<div class="shrink-0 self-center pr-2">
-					<Spinner className="size-3" />
-				</div>
-			{/if}
+					if ($selectedFolder) {
+						selectedFolder.set(null);
+					}
 
-			<div class="flex self-center flex-1 w-full min-w-0">
-				{#if unread}
-					<div class="shrink-0 self-center pr-2.5 flex transition-opacity duration-300">
-						<div class="size-1.5 bg-sky-500 rounded-full" />
+					if ($mobile) {
+						showSidebar.set(false);
+					}
+
+					// Optimistically mark as read in UI when clicked
+					unread = false;
+					lastReadAt = Date.now() / 1000;
+				}}
+				ondblclick={async (e) => {
+					if (readonly) return;
+					e.preventDefault();
+					e.stopPropagation();
+
+					doubleClicked = true;
+					renameHandler();
+				}}
+				draggable="false"
+			>
+				{#if ownerUserId}
+					<Tooltip content={ownerName || 'Unknown'}>
+						<img
+							src="/api/v1/users/{ownerUserId}/profile/image"
+							alt=""
+							class="size-3.5 rounded-full shrink-0 object-cover mr-1.5"
+						/>
+					</Tooltip>
+				{/if}
+
+				<!-- Loading spinner for active chat (left side) -->
+				{#if active}
+					<div class="shrink-0 self-center pr-2">
+						<Spinner className="size-3" />
 					</div>
 				{/if}
-				<div
-					dir="auto"
-					class="text-left self-center overflow-hidden w-full h-[20px] truncate {unread
-						? 'font-normal text-gray-800 dark:text-gray-200'
-						: ''} {showInlineActions && !readonly ? 'pr-12' : ''}"
-				>
-					{title}
-				</div>
-			</div>
 
-			<!-- Time ago indicator -->
-			{#if (updatedAt ?? createdAt) && !showInlineActions}
-				<div class="shrink-0 self-center text-[10px] text-gray-400 dark:text-gray-500 pl-2">
-					{formatTimeAgo(updatedAt ?? createdAt)}
+				<div class="flex self-center flex-1 w-full min-w-0">
+					{#if unread}
+						<div class="shrink-0 self-center pr-2.5 flex transition-opacity duration-300">
+							<div class="size-1.5 bg-sky-500 rounded-full"></div>
+						</div>
+					{/if}
+					<div
+						dir="auto"
+						class="text-left self-center overflow-hidden w-full h-[20px] truncate {unread
+							? 'font-normal text-gray-800 dark:text-gray-200'
+							: ''} {showInlineActions && !readonly ? 'pr-12' : ''}"
+					>
+						{title}
+					</div>
 				</div>
-			{/if}
-		</a>
+
+				<!-- Time ago indicator -->
+				{#if (updatedAt ?? createdAt) && !showInlineActions}
+					<div class="shrink-0 self-center text-[10px] text-gray-400 dark:text-gray-500 pl-2">
+						{formatTimeAgo((updatedAt ?? createdAt) as number)}
+					</div>
+				{/if}
+			</LinkPreview.Trigger>
+
+			<ChatHoverPreview
+				chatId={id}
+				title={chatTitle || title}
+				{openPreview}
+				side="right"
+				align="center"
+			/>
+		</LinkPreview.Root>
 	{/if}
 
 	<!-- svelte-ignore a11y-no-static-element-interactions -->

From aaf2834db758bfec69408ab4cabcf324965c221c Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 12:03:26 -0400
Subject: [PATCH 22/29] refac

---
 src/lib/components/layout/Sidebar/ChatHoverPreview.svelte | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/src/lib/components/layout/Sidebar/ChatHoverPreview.svelte b/src/lib/components/layout/Sidebar/ChatHoverPreview.svelte
index 1a45373a45..0d663479cc 100644
--- a/src/lib/components/layout/Sidebar/ChatHoverPreview.svelte
+++ b/src/lib/components/layout/Sidebar/ChatHoverPreview.svelte
@@ -6,7 +6,6 @@
 	import Messages from '$lib/components/chat/Messages.svelte';
 	import Spinner from '$lib/components/common/Spinner.svelte';
 	import { user } from '$lib/stores';
-	import { createMessagesList } from '$lib/utils';
 
 	const i18n: any = getContext('i18n');
 
@@ -21,7 +20,6 @@
 
 	let selectedModels = [''];
 	let history: any = null;
-	let previewMessages: any[] = [];
 	let previewReady = false;
 	let loading = false;
 	let error = '';
@@ -56,7 +54,6 @@
 		loading = true;
 		error = '';
 		previewReady = false;
-		previewMessages = [];
 		history = null;
 		selectedModels = [''];
 
@@ -68,11 +65,9 @@
 				? chat.chat.models
 				: [chat.chat.models ?? ''];
 			history = chat.chat.history;
-			previewMessages = createMessagesList(chat.chat.history, chat.chat.history.currentId);
 			previewReady = true;
 			scrollPreviewToBottom();
 		} else if (chat) {
-			previewMessages = [];
 			previewReady = true;
 		} else {
 			error = $i18n.t('Failed to load chat preview');
@@ -135,7 +130,7 @@
 						atSelectedModel={null}
 						{history}
 						autoScroll={true}
-						messagesCount={null}
+						messagesCount={8}
 						sendMessage={() => {}}
 						continueResponse={() => {}}
 						regenerateResponse={() => {}}

From 32242a678856a4e58b57f2c4b372efde36678312 Mon Sep 17 00:00:00 2001
From: Classic298 <27028174+Classic298@users.noreply.github.com>
Date: Thu, 23 Jul 2026 18:11:38 +0200
Subject: [PATCH 23/29] perf: 40% LESS CPU usage: cut per-instance CPU cost of
 shared socket.io Redis pub/sub channel (#27282)

* perf: cut per-instance CPU cost of shared socket.io Redis pub/sub channel

Profiling a multi-instance deployment (py-spy --gil) showed ~44% of worker
CPU in the socket.io pub/sub listener. Two causes, two fixes:

- Add hiredis so redis-py parses the RESP protocol in C instead of pure
  Python (redis/_parsers/resp3.py alone accounted for ~28% of GIL samples;
  redis-py auto-selects the hiredis parser when importable).

- Subclass AsyncRedisManager to drop emits whose target room has no local
  participants before upstream _handle_emit re-encodes the full packet.
  Every instance receives every emit published on the shared channel, so
  with N instances all but the hosting one were paying full packet
  re-serialization per message just to deliver it to nobody. Broadcasts
  (room=None) are unaffected.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q9CQ9qnp3sZGYQQwztsCJT

* fix: restrict pub/sub emit early-out to string rooms

Adversarial review against python-socketio 5.16.2 found one divergence
from upstream: for a degenerate empty-sequence room (emit to room=[]) on
an instance whose namespace has no local clients, the filter's
get_participants probe raises IndexError from room[0] where upstream
returns silently at the namespace guard and still publishes to Redis.
Open WebUI only ever emits to scalar string rooms or room=None, so the
case is unreachable today; guard on isinstance(room, str) anyway so any
non-string room shape passes through to upstream behavior unchanged.

Every open-webui emit uses a string room, so the fast path still covers
all real traffic.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q9CQ9qnp3sZGYQQwztsCJT

* Update requirements.txt

* Update pyproject.toml

* Update requirements-min.txt

---------

Co-authored-by: Claude <noreply@anthropic.com>
---
 backend/open_webui/socket/main.py | 25 ++++++++++++++++++++++++-
 backend/requirements-min.txt      |  1 +
 backend/requirements.txt          |  3 ++-
 pyproject.toml                    |  3 ++-
 4 files changed, 29 insertions(+), 3 deletions(-)

diff --git a/backend/open_webui/socket/main.py b/backend/open_webui/socket/main.py
index cb01d62e00..25792b0e78 100644
--- a/backend/open_webui/socket/main.py
+++ b/backend/open_webui/socket/main.py
@@ -57,6 +57,29 @@ REDIS = None
 # Configure CORS for Socket.IO
 SOCKETIO_CORS_ORIGINS = '*' if CORS_ALLOW_ORIGIN == ['*'] else CORS_ALLOW_ORIGIN
 
+class LocalFilteredRedisManager(socketio.AsyncRedisManager):
+    """AsyncRedisManager that drops pub/sub emits with no local recipients.
+
+    Every instance subscribed to the shared Socket.IO channel receives every
+    emit published by the whole fleet. Upstream ``_handle_emit`` re-encodes
+    the full packet before discovering the target room has no participants on
+    this instance, so each instance burns CPU serializing payloads addressed
+    to sessions it does not host — a cost that grows with instance count.
+    Bail out before that work when the room is empty here. Only string
+    rooms take the fast path; broadcasts (``room=None``) and any other
+    room shape (e.g. lists, which upstream indexes before validating)
+    always pass through unchanged.
+    """
+
+    async def _handle_emit(self, message):
+        room = message.get('room')
+        if isinstance(room, str):
+            namespace = message.get('namespace') or '/'
+            if next(self.get_participants(namespace, room), None) is None:
+                return
+        await super()._handle_emit(message)
+
+
 if WEBSOCKET_MANAGER == 'redis':
     sentinel_hosts = WEBSOCKET_SENTINEL_HOSTS or ''
     ws_redis_url = (
@@ -64,7 +87,7 @@ if WEBSOCKET_MANAGER == 'redis':
         if sentinel_hosts
         else WEBSOCKET_REDIS_URL
     )
-    redis_manager = socketio.AsyncRedisManager(ws_redis_url, redis_options=WEBSOCKET_REDIS_OPTIONS)
+    redis_manager = LocalFilteredRedisManager(ws_redis_url, redis_options=WEBSOCKET_REDIS_OPTIONS)
     sio = socketio.AsyncServer(
         cors_allowed_origins=SOCKETIO_CORS_ORIGINS,
         async_mode='asgi',
diff --git a/backend/requirements-min.txt b/backend/requirements-min.txt
index d5646443e6..7ccc3f7261 100644
--- a/backend/requirements-min.txt
+++ b/backend/requirements-min.txt
@@ -33,6 +33,7 @@ alembic==1.18.4
 
 pycrdt==0.13.1
 redis
+hiredis
 
 APScheduler==3.11.2
 RestrictedPython==8.2
diff --git a/backend/requirements.txt b/backend/requirements.txt
index e33bf131ea..4a3169762f 100644
--- a/backend/requirements.txt
+++ b/backend/requirements.txt
@@ -30,7 +30,8 @@ psycopg[binary]==3.3.4
 alembic==1.18.4
 
 pycrdt==0.13.1
-redis==8.0.0
+redis==8.0.1
+hiredis==3.4.0
 
 APScheduler==3.11.2
 RestrictedPython==8.2
diff --git a/pyproject.toml b/pyproject.toml
index 74659791df..1944904e53 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -38,7 +38,8 @@ dependencies = [
     "alembic==1.18.4",
 
     "pycrdt==0.13.1",
-    "redis==8.0.0",
+    "redis==8.0.1",
+    "hiredis==3.4.0",
     # "valkey-glide-sync==2.3.1",  # optional: install manually if VECTOR_DB=valkey
 
     "pytz==2026.2",

From 9a54bc4bbba636c68fd13d65ae2d3f2ff29eaac5 Mon Sep 17 00:00:00 2001
From: Classic298 <27028174+Classic298@users.noreply.github.com>
Date: Thu, 23 Jul 2026 18:11:51 +0200
Subject: [PATCH 24/29] perf: halve function-table queries in get_all_models
 (#27230)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

get_all_models ran four function-table queries: global actions, active
actions, global filters, active filters. Global functions are by
definition (type, is_active=True, is_global=True) — a subset of the
active set — so the global id sets can be derived from the active-rows
queries' is_global flag. Four queries become two, and each dropped
query returned full rows including every plugin's source code.

Also folds the mid-function 'models.default_metadata' read into the
Config.get_many already issued at the top of the function (one fewer
round trip; the existing `or {}` default handling is preserved).


Claude-Session: https://claude.ai/code/session_01MHg5zs1VBjvRWQ54qHpfYD

Co-authored-by: Claude <noreply@anthropic.com>
---
 backend/open_webui/utils/models.py | 33 +++++++++++++++---------------
 1 file changed, 16 insertions(+), 17 deletions(-)

diff --git a/backend/open_webui/utils/models.py b/backend/open_webui/utils/models.py
index 4012685965..9421975dcf 100644
--- a/backend/open_webui/utils/models.py
+++ b/backend/open_webui/utils/models.py
@@ -68,6 +68,7 @@ async def get_all_models(request, refresh: bool = False, user: UserModel = None)
         'models.base_models_cache',
         'evaluation.arena.enable',
         'evaluation.arena.models',
+        'models.default_metadata',
     )
     if (
         request.app.state.MODELS
@@ -125,23 +126,21 @@ async def get_all_models(request, refresh: bool = False, user: UserModel = None)
             ]
         models = models + arena_models
 
-    global_action_ids = (
-        {function.id for function in await Functions.get_global_action_functions()} if ENABLE_PLUGINS else set()
-    )
-    enabled_action_ids = (
-        {function.id for function in await Functions.get_functions_by_type('action', active_only=True)}
-        if ENABLE_PLUGINS
-        else set()
-    )
+    # One query per type: the global sets are subsets of the active sets, so
+    # deriving them from the same rows halves the function-table queries.
+    if ENABLE_PLUGINS:
+        active_actions = await Functions.get_functions_by_type('action', active_only=True)
+        global_action_ids = {function.id for function in active_actions if function.is_global}
+        enabled_action_ids = {function.id for function in active_actions}
 
-    global_filter_ids = (
-        {function.id for function in await Functions.get_global_filter_functions()} if ENABLE_PLUGINS else set()
-    )
-    enabled_filter_ids = (
-        {function.id for function in await Functions.get_functions_by_type('filter', active_only=True)}
-        if ENABLE_PLUGINS
-        else set()
-    )
+        active_filters = await Functions.get_functions_by_type('filter', active_only=True)
+        global_filter_ids = {function.id for function in active_filters if function.is_global}
+        enabled_filter_ids = {function.id for function in active_filters}
+    else:
+        global_action_ids = set()
+        enabled_action_ids = set()
+        global_filter_ids = set()
+        enabled_filter_ids = set()
 
     custom_models = await Models.get_all_models()
 
@@ -305,7 +304,7 @@ async def get_all_models(request, refresh: bool = False, user: UserModel = None)
 
     # Apply global model defaults to all models
     # Per-model overrides take precedence over global defaults
-    default_metadata = await Config.get('models.default_metadata', {}) or {}
+    default_metadata = config.get('models.default_metadata') or {}
 
     if default_metadata:
         for model in models:

From 9a6d16849968962ddbad6bbba548f722af2fa90b Mon Sep 17 00:00:00 2001
From: Classic298 <27028174+Classic298@users.noreply.github.com>
Date: Thu, 23 Jul 2026 18:23:05 +0200
Subject: [PATCH 25/29] fix: enforce action availability and model access on
 the chat action route (#27243)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The chat action route loaded a Function by its raw action_id and executed its action callable after only checking that the id and the requested model existed. The model list that the client renders actions from resolves each model's actions to the active action-type Functions that are global or assigned to that model, and the action route did not mirror that resolution, so a disabled, unassigned, or wrong-type Function, or an action on a model the caller cannot access, could be reached by calling the route directly.

Gate the route on the same rules the model resolution applies: the Function must be an active action, and for server-resolved models the caller must have model access and the action must be one the model actually surfaces (matched by function id, the prefix of each model actions entry, so single and sub-actions both resolve). Direct connections carry a client-supplied model the caller already owns, so the model-bound checks are scoped to non-direct calls; the active-action check always applies. Executing admin-authored Function code remains intended behaviour — this only keeps the route consistent with which actions each model exposes.

Co-authored-by: komyunghan <komyunghan@users.noreply.github.com>
---
 backend/open_webui/utils/actions.py | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/backend/open_webui/utils/actions.py b/backend/open_webui/utils/actions.py
index 585a409b72..1249dd60cb 100644
--- a/backend/open_webui/utils/actions.py
+++ b/backend/open_webui/utils/actions.py
@@ -9,7 +9,7 @@ from open_webui.models.functions import Functions
 from open_webui.models.users import UserModel
 from open_webui.socket.main import get_event_call, get_event_emitter
 from open_webui.utils.middleware import process_tool_result
-from open_webui.utils.models import get_all_models
+from open_webui.utils.models import check_model_access, get_all_models
 from open_webui.utils.plugin import get_function_module_from_cache
 
 logging.basicConfig(stream=sys.stdout, level=GLOBAL_LOG_LEVEL)
@@ -46,6 +46,23 @@ async def chat_action(request: Request, action_id: str, form_data: dict, user: A
         raise Exception('Model not found')
     model = models[model_id]
 
+    # Availability gate — keep this route consistent with the actions a model
+    # actually surfaces to the client. Executing admin-authored Function code is
+    # intended; this only stops a disabled, unassigned, or access-restricted
+    # action from being reached by calling the route with a raw action_id.
+    if action.type != 'action' or not action.is_active:
+        raise Exception(f'Action not available: {action_id}')
+
+    # Direct connections carry a client-supplied model the caller already owns,
+    # so scope the model-bound checks to server-resolved models.
+    if not getattr(request.state, 'direct', False) and user.role != 'admin':
+        await check_model_access(user, model)
+        # model['actions'] entries are '<function_id>' or '<function_id>.<sub_id>';
+        # the function id is always the prefix.
+        surfaced_action_ids = {item.get('id', '').split('.', 1)[0] for item in model.get('actions', [])}
+        if action_id not in surfaced_action_ids:
+            raise Exception(f'Action not available: {action_id}')
+
     __event_emitter__ = await get_event_emitter(
         {
             'chat_id': data['chat_id'],

From 8becf9443e3473179eee62085a682929c162d447 Mon Sep 17 00:00:00 2001
From: Classic298 <27028174+Classic298@users.noreply.github.com>
Date: Thu, 23 Jul 2026 18:23:24 +0200
Subject: [PATCH 26/29] fix: only admins may switch the global Automatic1111
 checkpoint from image generation (#27244)

The Automatic1111 branch of the image generation route called set_image_model whenever the request carried a model field. On this backend set_image_model is not request-scoped: it persists image_generation.model to the global configuration and posts the new sd_model_checkpoint to the shared server, because Automatic1111 holds a single checkpoint instance-wide. A non-admin holding features.image_generation could therefore change the instance-wide image model and the shared backend checkpoint for every user by sending a model on an ordinary generation request, even though the setting is otherwise managed only through the admin-only image configuration route and the frontend never sends this field.

Gate the switch on an admin caller. A non-admin now generates on the currently configured checkpoint and the model field no longer mutates global state; admins retain per-request model switching here and through the image configuration route. Image editing is unaffected, as it selects its model per request without writing global configuration.
---
 backend/open_webui/routers/images.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/backend/open_webui/routers/images.py b/backend/open_webui/routers/images.py
index 78247c0b7e..d26734da64 100644
--- a/backend/open_webui/routers/images.py
+++ b/backend/open_webui/routers/images.py
@@ -785,7 +785,11 @@ async def image_generations(
                 images.append({'url': url})
             return images
         elif image_config.IMAGE_GENERATION_ENGINE == 'automatic1111' or image_config.IMAGE_GENERATION_ENGINE == '':
-            if form_data.model:
+            # Automatic1111 holds one checkpoint instance-wide, so set_image_model
+            # persists the global default and switches the shared backend. Only an
+            # admin may do that; a non-admin generates on the currently configured
+            # checkpoint. The model field is not a per-user selection on this backend.
+            if form_data.model and user.role == 'admin':
                 await set_image_model(request, form_data.model)
 
             data = {

From 62491debfa26cfadb4ab445f2712172668212c33 Mon Sep 17 00:00:00 2001
From: Classic298 <27028174+Classic298@users.noreply.github.com>
Date: Thu, 23 Jul 2026 18:24:29 +0200
Subject: [PATCH 27/29] perf: linear content accumulation in the streaming
 response handler (#27231)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The streaming handler rebuilt the full accumulated response with
`content = f'{content}{value}'` on every content delta — a complete
string copy per chunk, making accumulation O(n^2) over the response
length. Use in-place `content += value` for the (universal) str case,
which CPython extends in place, keeping accumulation O(n); the
f-string fallback is preserved for non-str values so coercion
behavior is unchanged.

In the ENABLE_REALTIME_CHAT_SAVE branch, full_output() — which
concatenates the entire accumulated output — was called twice per
chunk (once for the DB upsert, once for the emitted delta). Compute
it once and reuse.


Claude-Session: https://claude.ai/code/session_01MHg5zs1VBjvRWQ54qHpfYD

Co-authored-by: Claude <noreply@anthropic.com>
---
 backend/open_webui/utils/middleware.py | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/backend/open_webui/utils/middleware.py b/backend/open_webui/utils/middleware.py
index a1a2a432f5..db86b0c8be 100644
--- a/backend/open_webui/utils/middleware.py
+++ b/backend/open_webui/utils/middleware.py
@@ -4490,7 +4490,12 @@ async def streaming_chat_response_handler(response, ctx):
                                                 user,
                                             )
 
-                                        content = f'{content}{value}'
+                                        if isinstance(value, str):
+                                            # In-place append — avoids copying the full
+                                            # accumulated response on every chunk.
+                                            content += value
+                                        else:
+                                            content = f'{content}{value}'
 
                                         # Check if we're inside a tag-based block
                                         # (reasoning, code_interpreter, or solution).
@@ -4597,16 +4602,17 @@ async def streaming_chat_response_handler(response, ctx):
                                         if ENABLE_REALTIME_CHAT_SAVE and not metadata.get('chat_id', '').startswith(
                                             'channel:'
                                         ):
+                                            current_output = full_output()
                                             # Save message in the database
                                             await Chats.upsert_message_to_chat_by_id_and_message_id(
                                                 metadata['chat_id'],
                                                 metadata['message_id'],
                                                 {
-                                                    'output': full_output(),
+                                                    'output': current_output,
                                                 },
                                             )
                                             data = {
-                                                'output': full_output(),
+                                                'output': current_output,
                                             }
                                             delta_type = 'content'
                                         else:

From 1513ddaf58fe18029086880461d7cad0649a699c Mon Sep 17 00:00:00 2001
From: Timothy Jaeryang Baek <tim@openwebui.com>
Date: Thu, 23 Jul 2026 12:25:17 -0400
Subject: [PATCH 28/29] refac

---
 src/lib/components/chat/Messages.svelte       | 41 ++++++++++++-------
 .../layout/Sidebar/ChatHoverPreview.svelte    |  4 ++
 2 files changed, 31 insertions(+), 14 deletions(-)

diff --git a/src/lib/components/chat/Messages.svelte b/src/lib/components/chat/Messages.svelte
index 14d22d5b66..740a8981fd 100644
--- a/src/lib/components/chat/Messages.svelte
+++ b/src/lib/components/chat/Messages.svelte
@@ -49,6 +49,7 @@
 	export let topPadding = false;
 	export let bottomPadding = false;
 	export let autoScroll;
+	export let messagesContainerId = 'messages-container';
 
 	export let onSelect = (e) => {};
 	export let onInsertToNote: ((content: string) => void) | null = null;
@@ -56,14 +57,18 @@
 	export let messagesCount: number | null = 8;
 	let messagesLoading = false;
 
+	const getMessagesContainer = () => document.getElementById(messagesContainerId);
+
 	onDestroy(() => {
 		cancelAnimationFrame(pendingRebuild);
 	});
 
 	const loadMoreMessages = async () => {
 		// scroll slightly down to disable continuous loading
-		const element = document.getElementById('messages-container');
-		element.scrollTop = element.scrollTop + 100;
+		const element = getMessagesContainer();
+		if (element) {
+			element.scrollTop = element.scrollTop + 100;
+		}
 
 		messagesLoading = true;
 		messagesCount += 8;
@@ -135,7 +140,7 @@
 	}
 
 	const scrollToBottom = () => {
-		const element = document.getElementById('messages-container');
+		const element = getMessagesContainer();
 		if (element) {
 			element.scrollTop = element.scrollHeight;
 
@@ -215,8 +220,10 @@
 
 		// Optional auto-scroll
 		if ($settings?.scrollOnBranchChange ?? true) {
-			const element = document.getElementById('messages-container');
-			autoScroll = element.scrollHeight - element.scrollTop <= element.clientHeight + 50;
+			const element = getMessagesContainer();
+			autoScroll = element
+				? element.scrollHeight - element.scrollTop <= element.clientHeight + 50
+				: false;
 
 			setTimeout(() => {
 				scrollToBottom();
@@ -262,8 +269,10 @@
 		await tick();
 
 		if ($settings?.scrollOnBranchChange ?? true) {
-			const element = document.getElementById('messages-container');
-			autoScroll = element.scrollHeight - element.scrollTop <= element.clientHeight + 50;
+			const element = getMessagesContainer();
+			autoScroll = element
+				? element.scrollHeight - element.scrollTop <= element.clientHeight + 50
+				: false;
 
 			setTimeout(() => {
 				scrollToBottom();
@@ -313,8 +322,10 @@
 		await tick();
 
 		if ($settings?.scrollOnBranchChange ?? true) {
-			const element = document.getElementById('messages-container');
-			autoScroll = element.scrollHeight - element.scrollTop <= element.clientHeight + 50;
+			const element = getMessagesContainer();
+			autoScroll = element
+				? element.scrollHeight - element.scrollTop <= element.clientHeight + 50
+				: false;
 
 			setTimeout(() => {
 				scrollToBottom();
@@ -485,11 +496,13 @@
 
 	const triggerScroll = () => {
 		if (autoScroll) {
-			const element = document.getElementById('messages-container');
-			autoScroll = element.scrollHeight - element.scrollTop <= element.clientHeight + 50;
-			setTimeout(() => {
-				scrollToBottom();
-			}, 100);
+			const element = getMessagesContainer();
+			if (element) {
+				autoScroll = element.scrollHeight - element.scrollTop <= element.clientHeight + 50;
+				setTimeout(() => {
+					scrollToBottom();
+				}, 100);
+			}
 		}
 	};
 </script>
diff --git a/src/lib/components/layout/Sidebar/ChatHoverPreview.svelte b/src/lib/components/layout/Sidebar/ChatHoverPreview.svelte
index 0d663479cc..e4f964b654 100644
--- a/src/lib/components/layout/Sidebar/ChatHoverPreview.svelte
+++ b/src/lib/components/layout/Sidebar/ChatHoverPreview.svelte
@@ -26,6 +26,8 @@
 	let requestedChatId = '';
 	let loadToken = 0;
 
+	$: messagesContainerId = `chat-hover-preview-messages-${chatId}`;
+
 	const scrollPreviewToBottom = async () => {
 		await tick();
 		requestAnimationFrame(() => {
@@ -103,6 +105,7 @@
 				</div>
 			</div>
 			<div
+				id={messagesContainerId}
 				bind:this={messagesContainerElement}
 				class="max-h-[min(15rem,calc(100vh-4rem))] overflow-y-auto bg-white scrollbar-hover @container dark:bg-gray-850"
 			>
@@ -130,6 +133,7 @@
 						atSelectedModel={null}
 						{history}
 						autoScroll={true}
+						{messagesContainerId}
 						messagesCount={8}
 						sendMessage={() => {}}
 						continueResponse={() => {}}

From 8ace4f0a8a599d0a4cce9c5db1b117701a00b275 Mon Sep 17 00:00:00 2001
From: Classic298 <27028174+Classic298@users.noreply.github.com>
Date: Thu, 23 Jul 2026 18:29:01 +0200
Subject: [PATCH 29/29] perf: stop recomputing static work in per-response ASGI
 middlewares (#27229)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

SecurityHeadersMiddleware called set_security_headers() on every
response — 14 os.environ.get lookups plus a regex validation per
configured header, for values that are static for the process
lifetime. Compute the header list once at construction; when no
security env vars are set, skip wrapping send entirely.

RedirectMiddleware decoded and parse_qs'd the query string of every
GET, though it only acts on /watch?v= and ?shared= URLs. Add a cheap
path/substring precheck first; a false positive just falls through to
the previous full parse, so no redirect behavior changes.

Verified byte-identical responses (status, Location, header values)
against the previous implementations across redirect, passthrough,
and no-env cases.


Claude-Session: https://claude.ai/code/session_01MHg5zs1VBjvRWQ54qHpfYD

Co-authored-by: Claude <noreply@anthropic.com>
---
 backend/open_webui/utils/asgi_middleware.py  | 10 +++++++++-
 backend/open_webui/utils/security_headers.py |  7 +++++--
 2 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/backend/open_webui/utils/asgi_middleware.py b/backend/open_webui/utils/asgi_middleware.py
index 89154b6103..7189bfcad8 100644
--- a/backend/open_webui/utils/asgi_middleware.py
+++ b/backend/open_webui/utils/asgi_middleware.py
@@ -227,7 +227,15 @@ class RedirectMiddleware:
             return
 
         path = scope.get('path', '')
-        query_string = scope.get('query_string', b'').decode('latin-1', errors='replace')
+        raw_query = scope.get('query_string', b'')
+        # This middleware only acts on /watch?v= and ?shared= URLs; skip the
+        # decode + parse_qs work for every other GET. (A false positive on the
+        # substring check just falls through to the full parse below.)
+        if not (path.endswith('/watch') or b'shared' in raw_query):
+            await self.app(scope, receive, send)
+            return
+
+        query_string = raw_query.decode('latin-1', errors='replace')
         query_params = parse_qs(query_string)
 
         redirect_params: dict[str, str] = {}
diff --git a/backend/open_webui/utils/security_headers.py b/backend/open_webui/utils/security_headers.py
index 090abeea33..c6cbbce2ba 100644
--- a/backend/open_webui/utils/security_headers.py
+++ b/backend/open_webui/utils/security_headers.py
@@ -15,16 +15,19 @@ class SecurityHeadersMiddleware:
 
     def __init__(self, app: ASGIApp) -> None:
         self.app = app
+        # Headers derive only from env vars, which are static for the process
+        # lifetime — compute them once instead of per response.
+        self._headers = list(set_security_headers().items())
 
     async def __call__(self, scope: Scope, receive: Receive, send: Send) -> None:
-        if scope['type'] != 'http':
+        if scope['type'] != 'http' or not self._headers:
             await self.app(scope, receive, send)
             return
 
         async def send_with_security_headers(message: Message) -> None:
             if message['type'] == 'http.response.start':
                 headers = MutableHeaders(scope=message)
-                for key, value in set_security_headers().items():
+                for key, value in self._headers:
                     headers[key] = value
             await send(message)