From f639769ca9b112e756c38a1f49797a9392c299d6 Mon Sep 17 00:00:00 2001
From: Ryan Crabbe <ryan@berri.ai>
Date: Wed, 15 Apr 2026 21:38:18 -0700
Subject: [PATCH] fix(proxy): use flat soft_budget field for virtual key
 projected-spend alerts
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The projected-spend alert in _update_key_cache read from
existing_spend_obj.litellm_budget_table["soft_budget"], but the nested
dict is never populated for virtual keys (the combined_view SQL maps
budget fields to flat top-level attributes instead). This made the
check dead code — it silently short-circuited on every request, and
when unblocked, crashed update_cache with a Pydantic ValidationError
because _get_projected_spend_over_limit returns a date object but
CallInfo.projected_exceeded_date expects str.

Fixes: read from the flat existing_spend_obj.soft_budget field that IS
populated, and stringify projected_exceeded_date.

Also marks team soft budget email alerts as enterprise in docs.

Closes #20324
---
 .../docs/proxy/ui_team_soft_budget_alerts.md     | 10 ++++++++++
 litellm/proxy/proxy_server.py                    | 16 +++++-----------
 2 files changed, 15 insertions(+), 11 deletions(-)

diff --git a/docs/my-website/docs/proxy/ui_team_soft_budget_alerts.md b/docs/my-website/docs/proxy/ui_team_soft_budget_alerts.md
index 17c42e57c9a..413457ccb82 100644
--- a/docs/my-website/docs/proxy/ui_team_soft_budget_alerts.md
+++ b/docs/my-website/docs/proxy/ui_team_soft_budget_alerts.md
@@ -2,6 +2,16 @@ import Image from '@theme/IdealImage';
 
 # Team Soft Budget Alerts
 
+:::info
+
+✨ This is an Enterprise feature. Email budget alerts require an enterprise license.
+
+[Enterprise Pricing](https://www.litellm.ai/#pricing)
+
+[Get free 7-day trial key](https://www.litellm.ai/enterprise#trial)
+
+:::
+
 Set a soft budget on a team and get email alerts when spending crosses the threshold — without blocking any requests.
 
 ## Overview
diff --git a/litellm/proxy/proxy_server.py b/litellm/proxy/proxy_server.py
index 9981c049c18..749551fb656 100644
--- a/litellm/proxy/proxy_server.py
+++ b/litellm/proxy/proxy_server.py
@@ -1870,26 +1870,20 @@ async def update_cache(  # noqa: PLR0915
         ## CHECK IF USER PROJECTED SPEND > SOFT LIMIT
         if (
             existing_spend_obj.soft_budget_cooldown is False
-            and existing_spend_obj.litellm_budget_table is not None
+            and existing_spend_obj.soft_budget is not None
             and (
                 _is_projected_spend_over_limit(
                     current_spend=new_spend,
-                    soft_budget_limit=existing_spend_obj.litellm_budget_table[
-                        "soft_budget"
-                    ],
+                    soft_budget_limit=existing_spend_obj.soft_budget,
                 )
                 is True
             )
         ):
             projected_spend, projected_exceeded_date = _get_projected_spend_over_limit(
                 current_spend=new_spend,
-                soft_budget_limit=existing_spend_obj.litellm_budget_table.get(
-                    "soft_budget", None
-                ),
+                soft_budget_limit=existing_spend_obj.soft_budget,
             )  # type: ignore
-            soft_limit = existing_spend_obj.litellm_budget_table.get(
-                "soft_budget", float("inf")
-            )
+            soft_limit = existing_spend_obj.soft_budget
             call_info = CallInfo(
                 token=existing_spend_obj.token or "",
                 spend=new_spend,
@@ -1897,7 +1891,7 @@ async def update_cache(  # noqa: PLR0915
                 max_budget=soft_limit,
                 user_id=existing_spend_obj.user_id,
                 projected_spend=projected_spend,
-                projected_exceeded_date=projected_exceeded_date,
+                projected_exceeded_date=str(projected_exceeded_date),
                 event_group=Litellm_EntityType.KEY,
             )
             # alert user