chore(deps): bump vulnerable dependencies

.circleci/config.yml

@@ -98,6 +98,19 @@ commands:
       - wait_for_service:
           url: tcp://localhost:5432
           timeout: "60"
+  start_redis:
+    description: "Start a redis container on port 6379 and wait until it accepts connections. Use this to isolate a job from the shared remote Redis so concurrent CI pipelines don't contend for pod locks or buffer keys."
+    steps:
+      - run:
+          name: Start Redis
+          command: |
+            docker run -d \
+              --name redis-cache \
+              -p 6379:6379 \
+              redis:7-alpine@sha256:7aec734b2bb298a1d769fd8729f13b8514a41bf90fcdd1f38ec52267fbaa8ee6
+      - wait_for_service:
+          url: tcp://localhost:6379
+          timeout: "60"
   setup_litellm_enterprise_pip:
     steps:
       - run:
@@ -1591,6 +1604,7 @@ jobs:
           command: |
             uv sync --frozen --all-groups --all-extras --python 3.12
       - start_postgres
+      - start_redis
       - attach_workspace:
           at: ~/project
       - run:
@@ -1600,15 +1614,18 @@ jobs:
             docker images | grep litellm-docker-database
       - run:
           name: Run Docker container
-          # intentionally give bad redis credentials here
-          # the OTEL test - should get this as a trace
+          # Point the proxy at the job-local Redis (start_redis) instead of the
+          # shared remote Redis. The Redis transaction buffer uses a single
+          # global pod-lock key (cronjob_lock:db_spend_update_job) and a single
+          # global buffer list (litellm_spend_update_buffer); sharing those
+          # across concurrent CI pipelines causes spend flushes to stall or
+          # land in the wrong DB, which is what makes this test flaky.
           command: |
             docker run -d \
               -p 4000:4000 \
               -e DATABASE_URL=postgresql://postgres:postgres@host.docker.internal:5432/circle_test \
-              -e REDIS_HOST=$REDIS_HOST \
-              -e REDIS_PASSWORD=$REDIS_PASSWORD \
-              -e REDIS_PORT=$REDIS_PORT \
+              -e REDIS_HOST=host.docker.internal \
+              -e REDIS_PORT=6379 \
               -e LITELLM_MASTER_KEY="sk-1234" \
               -e OPENAI_API_KEY=$OPENAI_API_KEY \
               -e LITELLM_LICENSE=$LITELLM_LICENSE \
@@ -1638,12 +1655,14 @@ jobs:
           command: |
             uv run --no-sync python -m pytest -vv tests/spend_tracking_tests -x --junitxml=test-results/junit.xml --durations=5
           no_output_timeout: 15m
-            # Clean up first container
       - run:
           name: Stop and remove first container
+          when: always
           command: |
             docker stop my-app
             docker rm my-app
+            docker stop redis-cache
+            docker rm redis-cache
 
   proxy_multi_instance_tests:
     machine:

.github/workflows/create-release.yml

@@ -110,6 +110,8 @@ jobs:
   create-branch:
     name: Create Release Branch
     needs: release
+    permissions:
+      contents: write
     uses: ./.github/workflows/create-release-branch.yml
     with:
       tag: ${{ inputs.tag }}

Dockerfile

@@ -1,9 +1,9 @@
 # Base image for building
-ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:a5a619c1793039dcf92f02178f37c94bb3d6001403716da59d6092dfe8d9b502
+ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:f26d42a15d09d9a643b231df929fa3cf609bedc58a728eb445be89a9d8d1da9f
 
 # Runtime image
-ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:a5a619c1793039dcf92f02178f37c94bb3d6001403716da59d6092dfe8d9b502
-ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.10.9@sha256:10902f58a1606787602f303954cea099626a4adb02acbac4c69920fe9d278f82
+ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:f26d42a15d09d9a643b231df929fa3cf609bedc58a728eb445be89a9d8d1da9f
+ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:733b4042187702f832f7fdecb3aff14a61b288c4ca37af188bb5715c1caebaf8
 
 FROM $UV_IMAGE AS uvbin
 
@@ -69,25 +69,14 @@ FROM $LITELLM_RUNTIME_IMAGE AS runtime
 USER root
 
 RUN apk add --no-cache bash openssl tzdata nodejs npm python3 libsndfile supervisor && \
-    npm install -g npm@11.12.1 tar@7.5.11 glob@11.1.0 @isaacs/brace-expansion@5.0.1 minimatch@10.2.4 diff@8.0.3 && \
+    npm install -g npm@11.12.1 tar@7.5.11 glob@13.0.6 @isaacs/brace-expansion@5.0.1 brace-expansion@5.0.5 minimatch@10.2.4 diff@8.0.3 picomatch@4.0.4 && \
     GLOBAL="$(npm root -g)" && \
-    find "$GLOBAL/npm" -type d -name "tar" -path "*/node_modules/tar" | while read d; do \
-        rm -rf "$d" && cp -rL "$GLOBAL/tar" "$d"; \
+    for pkg in tar glob @isaacs/brace-expansion brace-expansion minimatch diff picomatch; do \
+        name="${pkg##*/}"; \
+        find "$GLOBAL/npm" -type d -name "$name" -path "*/node_modules/$pkg" | while read d; do \
+            rm -rf "$d" && cp -rL "$GLOBAL/$pkg" "$d"; \
+        done; \
     done && \
-    find "$GLOBAL/npm" -type d -name "glob" -path "*/node_modules/glob" | while read d; do \
-        rm -rf "$d" && cp -rL "$GLOBAL/glob" "$d"; \
-    done && \
-    find "$GLOBAL/npm" -type d -name "brace-expansion" -path "*/node_modules/@isaacs/brace-expansion" | while read d; do \
-        rm -rf "$d" && cp -rL "$GLOBAL/@isaacs/brace-expansion" "$d"; \
-    done && \
-    find "$GLOBAL/npm" -type d -name "minimatch" -path "*/node_modules/minimatch" | while read d; do \
-        rm -rf "$d" && cp -rL "$GLOBAL/minimatch" "$d"; \
-    done && \
-    find "$GLOBAL/npm" -type d -name "diff" -path "*/node_modules/diff" | while read d; do \
-        rm -rf "$d" && cp -rL "$GLOBAL/diff" "$d"; \
-    done && \
-    find /usr/local/lib /usr/lib -path "*/node_modules/npm/package.json" -exec \
-        sed -i 's/"tar": "\^7\.5\.[0-9]*"/"tar": "^7.5.10"/g; s/"minimatch": "\^10\.[0-9.]*"/"minimatch": "^10.2.4"/g' {} + 2>/dev/null && \
     npm cache clean --force && \
     { apk del --no-cache npm 2>/dev/null || true; }
 

docker/Dockerfile.alpine

@@ -3,7 +3,7 @@ ARG LITELLM_BUILD_IMAGE=python:3.11-alpine@sha256:f07e2ace46f560f09a6eeec7b4913b
 
 # Runtime image
 ARG LITELLM_RUNTIME_IMAGE=python:3.11-alpine@sha256:f07e2ace46f560f09a6eeec7b4913b80ee99546e749ef82342a419a326620856
-ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.10.9@sha256:10902f58a1606787602f303954cea099626a4adb02acbac4c69920fe9d278f82
+ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:733b4042187702f832f7fdecb3aff14a61b288c4ca37af188bb5715c1caebaf8
 
 FROM $UV_IMAGE AS uvbin
 

docker/Dockerfile.database

@@ -1,9 +1,9 @@
 # Base image for building
-ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:a5a619c1793039dcf92f02178f37c94bb3d6001403716da59d6092dfe8d9b502
+ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:f26d42a15d09d9a643b231df929fa3cf609bedc58a728eb445be89a9d8d1da9f
 
 # Runtime image
-ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:a5a619c1793039dcf92f02178f37c94bb3d6001403716da59d6092dfe8d9b502
-ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.10.9@sha256:10902f58a1606787602f303954cea099626a4adb02acbac4c69920fe9d278f82
+ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:f26d42a15d09d9a643b231df929fa3cf609bedc58a728eb445be89a9d8d1da9f
+ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:733b4042187702f832f7fdecb3aff14a61b288c4ca37af188bb5715c1caebaf8
 
 FROM $UV_IMAGE AS uvbin
 
@@ -84,8 +84,6 @@ RUN apk add --no-cache bash openssl tzdata nodejs npm python3 libsndfile supervi
     find "$GLOBAL/npm" -type d -name "diff" -path "*/node_modules/diff" | while read d; do \
         rm -rf "$d" && cp -rL "$GLOBAL/diff" "$d"; \
     done && \
-    find /usr/local/lib /usr/lib -path "*/node_modules/npm/package.json" -exec \
-        sed -i 's/"tar": "\^7\.5\.[0-9]*"/"tar": "^7.5.10"/g; s/"minimatch": "\^10\.[0-9.]*"/"minimatch": "^10.2.4"/g' {} + 2>/dev/null && \
     npm cache clean --force && \
     { apk del --no-cache npm 2>/dev/null || true; }
 

docker/Dockerfile.dev

@@ -3,7 +3,7 @@ ARG LITELLM_BUILD_IMAGE=python:3.13-slim@sha256:739e7213785e88c0f702dcdc12c0973a
 
 # Runtime image
 ARG LITELLM_RUNTIME_IMAGE=python:3.13-slim@sha256:739e7213785e88c0f702dcdc12c0973afcbd606dbf021a589cab77d6b00b579d
-ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.10.9@sha256:10902f58a1606787602f303954cea099626a4adb02acbac4c69920fe9d278f82
+ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:733b4042187702f832f7fdecb3aff14a61b288c4ca37af188bb5715c1caebaf8
 
 FROM $UV_IMAGE AS uvbin
 

docker/Dockerfile.health_check

@@ -1,4 +1,4 @@
-ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.10.9@sha256:10902f58a1606787602f303954cea099626a4adb02acbac4c69920fe9d278f82
+ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:733b4042187702f832f7fdecb3aff14a61b288c4ca37af188bb5715c1caebaf8
 FROM $UV_IMAGE AS uvbin
 
 FROM python:3.13-slim@sha256:739e7213785e88c0f702dcdc12c0973afcbd606dbf021a589cab77d6b00b579d

docker/Dockerfile.non_root

@@ -1,8 +1,8 @@
 # Base images
-ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:a5a619c1793039dcf92f02178f37c94bb3d6001403716da59d6092dfe8d9b502
-ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:a5a619c1793039dcf92f02178f37c94bb3d6001403716da59d6092dfe8d9b502
+ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:f26d42a15d09d9a643b231df929fa3cf609bedc58a728eb445be89a9d8d1da9f
+ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:f26d42a15d09d9a643b231df929fa3cf609bedc58a728eb445be89a9d8d1da9f
 ARG PROXY_EXTRAS_SOURCE=published
-ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.10.9@sha256:10902f58a1606787602f303954cea099626a4adb02acbac4c69920fe9d278f82
+ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:733b4042187702f832f7fdecb3aff14a61b288c4ca37af188bb5715c1caebaf8
 
 FROM $UV_IMAGE AS uvbin
 

enterprise/pyproject.toml

@@ -19,7 +19,7 @@ requires = ["uv_build==0.10.7"]
 build-backend = "uv_build"
 
 [tool.uv]
-required-version = "==0.10.9"
+required-version = ">=0.10.9"
 
 [tool.uv.build-backend]
 module-root = ""

litellm-proxy-extras/pyproject.toml

@@ -19,7 +19,7 @@ requires = ["uv_build==0.10.7"]
 build-backend = "uv_build"
 
 [tool.uv]
-required-version = "==0.10.9"
+required-version = ">=0.10.9"
 
 [tool.uv.build-backend]
 module-root = ""

litellm/litellm_core_utils/logging_worker.py

@@ -370,11 +370,17 @@ async def stop(self) -> None:
         self._running_tasks.clear()
 
     async def flush(self) -> None:
-        """Flush the logging queue."""
+        """Flush the logging queue.
+
+        Waits until every enqueued task has completed. ``queue.join()`` blocks
+        on the queue's unfinished-task counter (decremented by ``task_done()``),
+        so it correctly handles items that have been dequeued but whose
+        callback hasn't finished yet — ``queue.empty()`` would return True in
+        that window and cause us to skip the wait.
+        """
         if self._queue is None:
             return
-        while not self._queue.empty():
-            await self._queue.join()
+        await self._queue.join()
 
     async def clear_queue(self):
         """

litellm/llms/vertex_ai/common_utils.py

@@ -229,13 +229,19 @@ def get_vertex_base_url(
 ) -> str:
     """
     Get the base URL for Vertex AI API calls.
+
+    - ``global`` uses the global control plane host.
+    - Multi-region geographies (e.g. ``us``, ``eu``) use ``aiplatform.{geo}.rep.googleapis.com``.
+    - Regional locations (e.g. ``us-central1``) use ``{region}-aiplatform.googleapis.com``.
     """
     if vertex_location == "global":
         return "https://aiplatform.googleapis.com"
     if vertex_location is None:
         raise ValueError("vertex_location is required")
     if not re.match(r"^[a-z][a-z0-9-]*$", vertex_location):
         raise ValueError("Invalid vertex_location format")
+    if "-" not in vertex_location:
+        return f"https://aiplatform.{vertex_location}.rep.googleapis.com"
     return f"https://{vertex_location}-aiplatform.googleapis.com"
 
 

litellm/proxy/_types.py

@@ -3907,7 +3907,7 @@ class OrganizationMemberUpdateResponse(MemberUpdateResponse):
 
 
 class TeamInfoResponseObjectTeamTable(LiteLLM_TeamTable):
-    team_member_budget_table: Optional[LiteLLM_BudgetTable] = None
+    team_member_budget_table: Optional[LiteLLM_BudgetTableFull] = None
     # Resources inherited from access groups (separate from direct assignments)
     access_group_models: Optional[List[str]] = None
     access_group_mcp_server_ids: Optional[List[str]] = None

litellm/proxy/common_utils/reset_budget_job.py

@@ -632,28 +632,35 @@ async def reset_budget_windows(self) -> None:
 
         now = datetime.utcnow()
 
+        # Note on raw SQL: prisma-client-python does not support null-filtering
+        # on `Json?` columns (no DbNull/JsonNull sentinel — see
+        # RobertCraigie/prisma-client-py#714). We use `query_raw` with
+        # `IS NOT NULL` so we don't materialize every key/team row on each
+        # tick of the reset job. Writes still go through the ORM.
+
         # --- Keys ---
         try:
-            all_keys = await self.prisma_client.db.litellm_verificationtoken.find_many(
-                where={"budget_limits": {"not": None}}  # type: ignore[arg-type]
+            key_rows = await self.prisma_client.db.query_raw(
+                'SELECT token, budget_limits FROM "LiteLLM_VerificationToken" '
+                "WHERE budget_limits IS NOT NULL"
             )
-            for key in all_keys:
-                raw = key.budget_limits  # type: ignore[attr-defined]
+            for row in key_rows:
+                raw = row["budget_limits"]
                 if not raw:
                     continue
                 windows: list = raw if isinstance(raw, list) else json.loads(raw)
                 changed = False
                 for window in windows:
                     counter_key = (
-                        f"spend:key:{key.token}:window:{window['budget_duration']}"
+                        f"spend:key:{row['token']}:window:{window['budget_duration']}"
                     )
                     if await ResetBudgetJob._reset_expired_window(
                         window, counter_key, spend_counter_cache, now
                     ):
                         changed = True
                 if changed:
                     await self.prisma_client.db.litellm_verificationtoken.update(
-                        where={"token": key.token},
+                        where={"token": row["token"]},
                         data={"budget_limits": json.dumps(windows)},  # type: ignore[arg-type]
                     )
         except Exception as e:
@@ -663,26 +670,25 @@ async def reset_budget_windows(self) -> None:
 
         # --- Teams ---
         try:
-            all_teams = await self.prisma_client.db.litellm_teamtable.find_many(
-                where={"budget_limits": {"not": None}}  # type: ignore[arg-type]
+            team_rows = await self.prisma_client.db.query_raw(
+                'SELECT team_id, budget_limits FROM "LiteLLM_TeamTable" '
+                "WHERE budget_limits IS NOT NULL"
             )
-            for team in all_teams:
-                raw = team.budget_limits  # type: ignore[attr-defined]
+            for row in team_rows:
+                raw = row["budget_limits"]
                 if not raw:
                     continue
                 windows = raw if isinstance(raw, list) else json.loads(raw)
                 changed = False
                 for window in windows:
-                    counter_key = (
-                        f"spend:team:{team.team_id}:window:{window['budget_duration']}"
-                    )
+                    counter_key = f"spend:team:{row['team_id']}:window:{window['budget_duration']}"
                     if await ResetBudgetJob._reset_expired_window(
                         window, counter_key, spend_counter_cache, now
                     ):
                         changed = True
                 if changed:
                     await self.prisma_client.db.litellm_teamtable.update(
-                        where={"team_id": team.team_id},
+                        where={"team_id": row["team_id"]},
                         data={"budget_limits": json.dumps(windows)},  # type: ignore[arg-type]
                     )
         except Exception as e: