diff --git a/tests/test_litellm/test_chat_ui_responses_session.py b/tests/test_litellm/test_chat_ui_responses_session.py
index 77ee1625864..2ecd7050e21 100644
--- a/tests/test_litellm/test_chat_ui_responses_session.py
+++ b/tests/test_litellm/test_chat_ui_responses_session.py
@@ -1,13 +1,13 @@
"""
-Tests for the chat UI responses API session chaining logic.
+Tests for responses API session chaining used by the chat UI.
-Validates that:
-1. previous_response_id is correctly forwarded in responses API calls
-2. The parameter is omitted (not sent as None) when starting a new session
+Verifies that:
+1. previous_response_id is correctly forwarded when provided
+2. Absence of previous_response_id does not break the call
+3. The aresponses function signature exposes the expected parameters
"""
import os
import sys
-from unittest.mock import AsyncMock, MagicMock, patch
sys.path.insert(0, os.path.abspath("../.."))
@@ -19,52 +19,118 @@ import litellm
class TestResponsesSessionChaining:
"""Test previous_response_id session chaining for the chat UI."""
- @pytest.mark.asyncio
- async def test_previous_response_id_forwarded(self):
- """previous_response_id should be passed through to the responses API call."""
- captured = {}
-
- async def fake_aresponses(*args, **kwargs):
- captured.update(kwargs)
- mock_resp = MagicMock()
- mock_resp.id = "resp_abc123"
- mock_resp.output = []
- return mock_resp
-
- with patch("litellm.aresponses", side_effect=fake_aresponses):
- await litellm.aresponses(
- input="Hello",
- model="gpt-4o",
- previous_response_id="resp_prev999",
- )
-
- assert captured.get("previous_response_id") == "resp_prev999"
-
- @pytest.mark.asyncio
- async def test_new_session_has_no_previous_response_id(self):
- """A new conversation should not send previous_response_id."""
- captured = {}
-
- async def fake_aresponses(*args, **kwargs):
- captured.update(kwargs)
- mock_resp = MagicMock()
- mock_resp.id = "resp_new001"
- mock_resp.output = []
- return mock_resp
-
- with patch("litellm.aresponses", side_effect=fake_aresponses):
- await litellm.aresponses(
- input="Hello",
- model="gpt-4o",
- # No previous_response_id — new session
- )
-
- assert captured.get("previous_response_id") is None
-
def test_responses_api_signature_accepts_previous_response_id(self):
- """Smoke test: aresponses function signature accepts previous_response_id."""
+ """aresponses must accept previous_response_id and onResponseId-like params."""
import inspect
+
sig = inspect.signature(litellm.aresponses)
assert "previous_response_id" in sig.parameters, (
- "aresponses must accept previous_response_id for session chaining"
+ "aresponses must accept previous_response_id for multi-turn session chaining"
+ )
+ assert "input" in sig.parameters, "aresponses must accept input"
+ assert "model" in sig.parameters, "aresponses must accept model"
+
+ @pytest.mark.asyncio
+ async def test_previous_response_id_included_in_request_body(self):
+ """previous_response_id must appear in the outgoing HTTP request body."""
+ import httpx
+
+ captured_body: dict = {}
+
+ async def mock_send(self_transport, request: httpx.Request, **kwargs):
+ import json
+
+ try:
+ captured_body.update(json.loads(request.content))
+ except Exception:
+ pass
+ # Return a minimal valid responses API response
+ response_json = {
+ "id": "resp_test123",
+ "object": "response",
+ "model": "gpt-4o-mini",
+ "output": [
+ {
+ "type": "message",
+ "id": "msg_001",
+ "role": "assistant",
+ "content": [{"type": "output_text", "text": "hi", "annotations": []}],
+ "status": "completed",
+ }
+ ],
+ "usage": {"input_tokens": 5, "output_tokens": 3, "total_tokens": 8},
+ "status": "completed",
+ "created_at": 1700000000,
+ }
+ return httpx.Response(
+ 200,
+ json=response_json,
+ request=request,
+ )
+
+ import unittest.mock as mock
+
+ with mock.patch("httpx.AsyncClient.send", mock_send):
+ try:
+ await litellm.aresponses(
+ input="hello",
+ model="gpt-4o-mini",
+ previous_response_id="resp_prev_abc",
+ api_key="sk-test-fake",
+ )
+ except Exception:
+ pass # response parsing may fail; we only care about the outgoing body
+
+ assert captured_body.get("previous_response_id") == "resp_prev_abc", (
+ f"Expected previous_response_id in request body, got: {captured_body}"
+ )
+
+ @pytest.mark.asyncio
+ async def test_no_previous_response_id_omitted_from_request(self):
+ """When previous_response_id is None, it must not appear in the request body."""
+ import httpx
+
+ captured_body: dict = {}
+
+ async def mock_send(self_transport, request: httpx.Request, **kwargs):
+ import json
+
+ try:
+ captured_body.update(json.loads(request.content))
+ except Exception:
+ pass
+ response_json = {
+ "id": "resp_new001",
+ "object": "response",
+ "model": "gpt-4o-mini",
+ "output": [
+ {
+ "type": "message",
+ "id": "msg_001",
+ "role": "assistant",
+ "content": [{"type": "output_text", "text": "hi", "annotations": []}],
+ "status": "completed",
+ }
+ ],
+ "usage": {"input_tokens": 5, "output_tokens": 3, "total_tokens": 8},
+ "status": "completed",
+ "created_at": 1700000000,
+ }
+ return httpx.Response(200, json=response_json, request=request)
+
+ import unittest.mock as mock
+
+ with mock.patch("httpx.AsyncClient.send", mock_send):
+ try:
+ await litellm.aresponses(
+ input="hello",
+ model="gpt-4o-mini",
+ previous_response_id=None,
+ api_key="sk-test-fake",
+ )
+ except Exception:
+ pass
+
+ assert "previous_response_id" not in captured_body, (
+ "previous_response_id must be omitted from the request body when None"
)
diff --git a/ui/litellm-dashboard/src/app/(dashboard)/components/Sidebar2.tsx b/ui/litellm-dashboard/src/app/(dashboard)/components/Sidebar2.tsx
index 9c2b47946a5..ceb14864ad9 100644
--- a/ui/litellm-dashboard/src/app/(dashboard)/components/Sidebar2.tsx
+++ b/ui/litellm-dashboard/src/app/(dashboard)/components/Sidebar2.tsx
@@ -160,7 +160,6 @@ const toHref = (slugOrPath: string) => {
// ----- Menu config (unchanged labels/icons; same appearance) -----
const menuItems: MenuItemCfg[] = [
{ key: "1", page: "api-keys", label: "Virtual Keys", icon: },
- { key: "29", page: "chat", label: "Chat", icon: , newTab: true },
{
key: "3",
page: "llm-playground",
@@ -424,6 +423,8 @@ const Sidebar2: React.FC = ({ accessToken, userRole, defaultSelect
style={{
transition: "all 0.3s cubic-bezier(0.4, 0, 0.2, 1)",
position: "relative",
+ display: "flex",
+ flexDirection: "column",
}}
>
= ({ accessToken, userRole, defaultSelect
borderRight: 0,
backgroundColor: "transparent",
fontSize: "14px",
+ flex: 1,
+ overflowY: "auto",
}}
items={filteredMenuItems.map((item) => ({
key: item.key,
diff --git a/ui/litellm-dashboard/src/components/chat/ChatPage.tsx b/ui/litellm-dashboard/src/components/chat/ChatPage.tsx
index af235665f59..bd21926fd3d 100644
--- a/ui/litellm-dashboard/src/components/chat/ChatPage.tsx
+++ b/ui/litellm-dashboard/src/components/chat/ChatPage.tsx
@@ -27,6 +27,7 @@ import MCPAppsPanel from "./MCPAppsPanel";
import { fetchAvailableModels } from "../playground/llm_calls/fetch_models";
import { makeOpenAIChatCompletionRequest } from "../playground/llm_calls/chat_completion";
import { makeOpenAIResponsesRequest } from "../playground/llm_calls/responses_api";
+import MCPEventsDisplay, { MCPEvent } from "../playground/chat_ui/MCPEventsDisplay";
import { getProxyBaseUrl } from "@/components/networking";
import { useUIConfig } from "@/app/(dashboard)/hooks/uiConfig/useUIConfig";
import { getProviderLogoAndName } from "@/components/provider_info_helpers";
@@ -137,6 +138,7 @@ const ChatPage: React.FC = ({ accessToken, userRole, userId, user
const [selectedMCPServers, setSelectedMCPServers] = useState([]);
const [responsesSessionId, setResponsesSessionId] = useState(null);
+ const [mcpEvents, setMcpEvents] = useState([]);
const [isStreaming, setIsStreaming] = useState(false);
const [inputText, setInputText] = useState("");
const [mcpPopoverOpen, setMcpPopoverOpen] = useState(false);
@@ -205,6 +207,12 @@ const ChatPage: React.FC = ({ accessToken, userRole, userId, user
if (staleId) router.replace(getChatUrl(uiRoot));
}, [staleId, router]);
+ // Reset the responses session when switching between conversations so that
+ // previous_response_id from conversation A is never sent for conversation B.
+ useEffect(() => {
+ setResponsesSessionId(null);
+ }, [activeConversationId]);
+
const toggleModel = useCallback((model: string) => {
setSelectedModels((prev) => {
let next: string[];
@@ -240,14 +248,19 @@ const ChatPage: React.FC = ({ accessToken, userRole, userId, user
appendMessage(convId, { role: "user", content: trimmed });
appendMessage(convId, { role: "assistant", content: "" });
+ setMcpEvents([]); // clear MCP events from previous request
setIsStreaming(true);
abortControllerRef.current = new AbortController();
+ // Explicitly filter to only user/assistant roles — tool messages lack
+ // a required tool_call_id and would cause API errors if forwarded.
const history = [
...(historyOverride ?? (activeConversation?.messages ?? [])
- .filter((m) => m.role === "user" || m.role === "assistant")
+ .filter((m): m is typeof m & { role: "user" | "assistant" } =>
+ m.role === "user" || m.role === "assistant"
+ )
.map((m) => ({
- role: m.role as "user" | "assistant",
+ role: m.role,
content: m.content,
}))),
{ role: "user" as const, content: trimmed },
@@ -275,6 +288,7 @@ const ChatPage: React.FC = ({ accessToken, userRole, userId, user
selectedMCPServers.length > 0 ? selectedMCPServers : undefined,
responsesSessionId,
(id: string) => setResponsesSessionId(id),
+ (event: MCPEvent) => setMcpEvents((prev) => [...prev, event]),
);
} catch (err: unknown) {
if (err instanceof Error && err.name === "AbortError") {
@@ -1120,11 +1134,18 @@ const ChatPage: React.FC = ({ accessToken, userRole, userId, user
})}
) : (
-
+ <>
+
+ {mcpEvents.length > 0 && (
+
+
+
+ )}
+
)}
{showScrollButton && (