fix(mcp): persist Tools-tab MCP OAuth token to DB by tin-berri · Pull Request #29809 · BerriAI/litellm

tin-berri · 2026-06-05T20:00:54Z

Problem

MCP OAuth "on behalf of" (OBO) mode was not persisting per-user access tokens, so tools came back empty and the connection never showed as established.
access_token and refresh_token were not stored in the database and never refreshed if they are expired

Fix

Stored access_token and refresh_token into the litellm_MCPUserCredentials DB Table
Added logic for minting new access_token if access_token is expired or has 60s left to expire

After Fix: Users are able to store access_tokens with LiteLLM and can access tools list consistently via a DB lookup for the corresponding access_tokens

Linear ticket

Resolves LIT-3579
https://linear.app/litellm-ai/issue/LIT-3579/fix-obo-mcp-tool-access-and-token-persistence

Note

Need to address the token_validation on the Create MCP Server Temporary Server Authentication since we are trying to move into simplifying the MCP Server Setup and in OBO-mode.

Pre-Submission checklist

Please complete all items before asking a LiteLLM maintainer to review your PR

I have added meaningful tests
My PR passes all unit tests on make test-unit
My PR's scope is as isolated as possible; it only solves 1 specific problem
I have requested a Greptile review by commenting @greptileai and received a Confidence Score of at least 4/5 before requesting a maintainer review

CI (LiteLLM team)

Branch creation CI run
Link:
CI run for the last commit
Link:
Merge / cherry-pick CI run
Links:

Screenshots / Proof of Fix

Deployed the branch to a vcluster and added a Slack MCP server in OBO mode (Transport HTTP, Auth OAuth, OAuth Flow Type Interactive, delegate_auth_to_upstream off), clicked "Authorize & Fetch", then submitted "Add MCP Server". Verified the token is written to the DB on submit and that the proxy can use it.

Prereqs: proxy reachable at localhost:8000 (port-forward), master key sk-compare-litellm, and the new server's id in SERVER_ID.

SERVER_ID=3c6789d7-8e47-4271-9343-545a485de2eb   # the server you just created

DB: the per-user OAuth credential row is created on submit (before the flow this is ROW_COUNT= 0)

GW=$(kubectl -n litellm get pod -o name | grep gateway | head -1 | cut -d/ -f2)
kubectl -n litellm exec -i "$GW" -- /app/.venv/bin/python - <<'PY'
import os, asyncio
from litellm.proxy.auth.rds_iam_token import generate_iam_auth_token
host=os.environ["DATABASE_HOST"]; port=os.getenv("DATABASE_PORT","5432")
user=os.getenv("DATABASE_USER") or "postgres"; name=os.getenv("DATABASE_NAME","postgres")
os.environ["DATABASE_URL"]=f"postgresql://{user}:{generate_iam_auth_token(db_host=host,db_port=port,db_user=user)}@{host}:{port}/{name}"
from prisma import Prisma
async def main():
    db=Prisma(); await db.connect()
    rows=await db.query_raw('SELECT user_id, server_id, created_at, length(credential_b64) AS cred_len FROM "LiteLLM_MCPUserCredentials" ORDER BY created_at DESC')
    print("ROW_COUNT=", len(rows))
    for r in rows: print(r)
    await db.disconnect()
asyncio.run(main())
PY

ROW_COUNT= 1
{'user_id': 'default_user_id', 'server_id': '3c6789d7-8e47-4271-9343-545a485de2eb', 'created_at': '2026-06-05T23:32:23.880+00:00', 'cred_len': 1232}

Proxy reads the stored credential (status endpoint)

curl -s -H "Authorization: Bearer sk-compare-litellm" \
  "http://localhost:8000/v1/mcp/server/$SERVER_ID/oauth-user-credential/status"

{"server_id":"3c6789d7-8e47-4271-9343-545a485de2eb","has_credential":true,"expires_at":"2026-06-06T11:32:23.867530+00:00","is_expired":false,"connected_at":"2026-06-05T23:32:23.867549+00:00"}

Tool list works with only the LiteLLM key, so the backend attaches the stored token (before the fix this returned {"tools": 0, ...} / "No tools available")

curl -s -H "Authorization: Bearer sk-compare-litellm" \
  "http://localhost:8000/mcp-rest/tools/list?server_id=$SERVER_ID" \
  | python3 -c "import sys,json;d=json.load(sys.stdin);print(json.dumps({'tools':len(d.get('tools') or []),'error':d.get('error'),'message':d.get('message')}))"

{"tools": 19, "error": null, "message": "Successfully retrieved tools"}

The DB check runs inside the gateway pod because the RDS connection uses IAM auth (no static password); it reuses the proxy's own generate_iam_auth_token so it connects exactly like the proxy. Adjust the namespace / gateway pod selector if your release differs.

Token refresh on expiry (the refresh half)

Same vcluster and the same Slack OBO server. The Tools-tab list path (/mcp-rest/tools/list) now refreshes a stored token that has expired, minting a new one from the stored refresh_token instead of returning an empty tool list.

Forcing expiry without waiting ~12h for Slack's natural token lifetime needs a DB write (the credential is encrypted, so it can't be done with curl). I rewrote the stored credential's expires_at into the past and recorded fingerprints (sha256 prefix, never the raw token), without calling any refresh code:

access_token_fp : 69e4f40e0f2b
refresh_token_fp: 846dbefb19f9
expires_at      : 2026-06-06T02:44:10Z   (in the past)

Then I opened the slack_OBO server's tools in the dashboard, which is the action that calls /mcp-rest/tools/list. The tools rendered, and the stored credential had been refreshed end to end with no manual refresh:

                  before (armed)      after opening the Tools tab
access_token_fp   69e4f40e0f2b   ->   b2d9372f0de4    rotated
refresh_token_fp  846dbefb19f9   ->   beddfed984e1    rotated (Slack rotation)
expires_at        02:44Z (past)  ->   15:46Z (~12h)   fresh

The access token changing proves a new token was minted from the stored refresh_token; the refresh token also rotating confirms Slack processed a real grant_type=refresh_token exchange rather than anything local. Before this change the same flow returned an empty tool list once the token lapsed

Type

🐛 Bug Fix

Changes

When adding an MCP server, "Authorize & Fetch" runs the OAuth code exchange and useMcpOAuthFlow holds the resulting access and refresh token. On submit, after createMCPServer returns the new server_id, the create flow only cached the token in sessionStorage (setToken) with no backend DB write. For an interactive PKCE oauth2 server with delegate_auth_to_upstream off (OBO mode) LiteLLM owns the per-user token, so skipping the DB write left has_credentials false; _get_user_oauth_extra_headers_from_db then returned None, a tools/list issued with only the LiteLLM key went upstream unauthenticated and came back empty, and the user had to re-authorize every time to see the tool list.

This change persists the token on submit for OBO servers via storeMCPOAuthUserCredential (POST /v1/mcp/server/{id}/oauth-user-credential), keyed by the calling user, so has_credentials becomes true and the proxy attaches the upstream auth header on subsequent tool lists. The mode is classified with getMcpOAuthMode, so only OBO writes to the DB; PKCE passthrough keeps its browser-held sessionStorage token (forwarded as the x-mcp-<alias>-authorization header) and M2M uses the backend service token.

Added a regression test in create_mcp_server.test.tsx asserting an OBO submit persists the access and refresh token to the DB via storeMCPOAuthUserCredential and does not fall back to the sessionStorage cache.

This PR also covers the second half of the OBO lifecycle: refreshing the stored token once it expires. The Tools-tab list endpoint (/mcp-rest/tools/list) read the stored credential and, when the access token had expired, returned None without attempting a refresh, so the tool list went empty again once the token lapsed even though a valid refresh_token was sitting in the DB. The MCP-protocol path in server.py already refreshed, so the two surfaces disagreed. The refresh decision (return the stored token while it is valid, otherwise mint a new one from the stored refresh_token) is now extracted into a shared resolve_valid_user_oauth_token helper in db.py, and both the REST and protocol paths call it, deleting server.py's duplicated inline refresh block. The helper also treats a token expiring within MCP_PER_USER_TOKEN_EXPIRY_BUFFER_SECONDS (60s) as needing refresh, so a near-expiry token is renewed before it can lapse mid-request; that buffer was previously applied only to the Redis cache TTL and not to the DB expiry check. Scope stays limited to OBO, the only mode that persists a per-user refresh_token; M2M caches its token in-memory per-server and passthrough keeps the credential browser-side, so neither has a stored credential for the helper to refresh.

Added regression tests in test_db_credentials.py covering the expiry buffer boundary (a token with 30s of life left is treated as valid under no buffer but as expired under the 60s buffer) and the refresh decision (expired-with-refresh-token mints a new token, expired-without-refresh-token returns None, a still-valid token is returned without any refresh call, and a failed refresh surfaces as None).

greptile-apps · 2026-06-05T20:03:37Z

Greptile Summary

This PR fixes MCP OAuth "on behalf of" mode on the Tools screen by adding a server-side credential persist step (storeMCPOAuthUserCredential) after the token exchange, matching the existing pattern in useUserMcpOAuthFlow. Previously, the token was cached only in sessionStorage, so has_credentials remained false and tool listings came back empty after a page refresh.

useToolsOAuthFlow.tsx: Inserts await storeMCPOAuthUserCredential(...) between the token exchange and the setToken / onSuccess calls; a DB write failure now propagates as an error and suppresses the false-success path (setToken and onSuccess are only reached when the write succeeds).
useToolsOAuthFlow.test.tsx: New unit tests verify the happy path (DB write, sessionStorage cache, and onSuccess all fire) and the failure path (status flips to error, setToken and onSuccess are not called).

Confidence Score: 4/5

Safe to merge — the change is minimal and surgical, touching only the Tools-screen OAuth hook. The core logic mirrors the already-reviewed useUserMcpOAuthFlow pattern, and both the happy path and the DB-failure path are covered by the new tests.

The implementation is clean and consistent with the established pattern. The one small gap is that the has_credential field in the successful response from storeMCPOAuthUserCredential is not checked — a 200 OK with has_credential: false would be treated as success. This is unlikely in practice (the server would return a non-OK status on actual failure) and is the same approach taken by useUserMcpOAuthFlow, so no regression is introduced. No other functional concerns were identified.

No files require special attention. useToolsOAuthFlow.tsx carries the only functional change and is straightforward to audit.

Important Files Changed

Filename	Overview
ui/litellm-dashboard/src/hooks/useToolsOAuthFlow.tsx	Adds a mandatory storeMCPOAuthUserCredential DB write after token exchange, mirroring useUserMcpOAuthFlow; setToken (sessionStorage) is only reached on successful DB write, so a failed write now surfaces as an error instead of a silent false-success.
ui/litellm-dashboard/src/hooks/useToolsOAuthFlow.test.tsx	New test file covering both the persist-on-success path (DB write called, setToken called, onSuccess called) and the no-false-success-on-DB-failure path (status flips to error, setToken and onSuccess not called); all network calls are properly mocked.

_{Reviews (1): Last reviewed commit: "fix(ui): persist Tools-tab MCP OAuth tok..." | Re-trigger Greptile}

codecov · 2026-06-05T20:04:28Z

Codecov Report

❌ Patch coverage is 75.00000% with 6 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
litellm/proxy/_experimental/mcp_server/db.py	88.23%	2 Missing ⚠️
...m/proxy/_experimental/mcp_server/rest_endpoints.py	0.00%	2 Missing ⚠️
litellm/proxy/_experimental/mcp_server/server.py	60.00%	2 Missing ⚠️

📢 Thoughts on this report? Let us know!

veria-ai · 2026-06-05T23:45:51Z

+          });
+          if (oauthMode === "obo") {
+            const scope = oauthTokenResponse.scope;
+            await storeMCPOAuthUserCredential(accessToken, response.server_id, {


Medium: OAuth token validation bypass

This stores the token returned by the temporary OAuth session, but getTemporaryPayload() does not include token_validation, and the /oauth-user-credential store endpoint just persists the posted token. A user creating or submitting an OBO MCP server can authorize an account that does not satisfy the configured token validation rules, then have that token accepted for later tool calls. Include the validation rules in the temporary OAuth session and/or have the credential storage path re-validate the token response against the persisted server before upserting it.

Confirmed, and it is actually broader than the temporary session. token_validation is not wired through to enforcement anywhere today: a grep across litellm/ finds it only on the in-memory MCPServer type and the _validate_token_response check at discoverable_endpoints.py:476, and nothing ever assigns mcp_server.token_validation in either manager load path. There is no DB column and NewMCPServerRequest has no token_validation field, so the rules entered in the UI are dropped before they could reach the check, for persisted servers too, not just the temporary session.

Wiring it end to end means a DB migration plus the persist path, manager population on load, and carrying it into the temporary record so the create-flow exchange validates. That is a bigger change than this PR, which is scoped to persisting and refreshing the OBO token, so I am handling token_validation in a dedicated follow-up PR and tracking it separately rather than resolving here

veria-ai · 2026-06-05T23:45:59Z

PR overview

This PR updates the Tools-tab MCP server creation flow so OAuth tokens obtained during the temporary authorization session are persisted to the database for later MCP tool use. The touched UI code is in the MCP server creation component that handles OAuth credential submission.

There is one open security issue in the OAuth credential persistence path. The current flow can store an OAuth token without enforcing the MCP server’s configured token validation rules, allowing a user to save a token for an account that should not be accepted and use it in later tool calls. No issues have been addressed yet, so the PR still needs validation added before this flow is safe to merge.

Open issues (1)

Medium: OAuth token validation bypass — ui/litellm-dashboard/src/components/mcp_tools/create_mcp_server.tsx:445

Fixed/addressed: 0 · PR risk: 7/10

…itellm_mcp_tools_obomode # Conflicts: # tests/test_litellm/proxy/_experimental/mcp_server/test_db_credentials.py

mateo-berri

LGTM; thanks!

The #29809 and #29960 cherry-picks landed onto v1.88.0, which predates the per-user env-var rotation feature (#28917) and the MCP upstream-auth exception module (#28356). Git's conflict resolution bundled those intermediate-commit tests into the picked test files, where they fail against this base because the code under test does not exist here (encrypt_value_helper, the env-var rotation helper, and litellm.proxy._experimental.mcp_server.exceptions.MCPUpstreamAuthError). Remove the leaked tests; the genuine regression tests added by #29809 (OBO refresh-on-expiry) and #29960 (admin-only include_disabled_tools) stay and pass.

…9.0) (#93) This PR contains the following updates: | Package | Update | Change | |---|---|---| | [ghcr.io/berriai/litellm](https://images.chainguard.dev/directory/image/wolfi-base/overview) ([source](https://github.com/BerriAI/litellm)) | minor | `v1.88.1` → `v1.89.0` | --- ### Release Notes <details> <summary>BerriAI/litellm (ghcr.io/berriai/litellm)</summary> ### [`v1.89.0`](https://github.com/BerriAI/litellm/releases/tag/v1.89.0) [Compare Source](https://github.com/BerriAI/litellm/compare/v1.89.0...v1.89.0) ##### Verify Docker Image Signature All LiteLLM Docker images are signed with [cosign](https://docs.sigstore.dev/cosign/overview/). Every release is signed with the same key introduced in [commit `0112e53`](https://github.com/BerriAI/litellm/commit/0112e53046018d726492c814b3644b7d376029d0). **Verify using the pinned commit hash (recommended):** A commit hash is cryptographically immutable, so this is the strongest way to ensure you are using the original signing key: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/0112e53046018d726492c814b3644b7d376029d0/cosign.pub \ ghcr.io/berriai/litellm:v1.89.0 ``` **Verify using the release tag (convenience):** Tags are protected in this repository and resolve to the same key. This option is easier to read but relies on tag protection rules: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/v1.89.0/cosign.pub \ ghcr.io/berriai/litellm:v1.89.0 ``` Expected output: ``` The following checks were performed on each of these signatures: - The cosign claims were validated - The signatures were verified against the specified public key ``` *** ##### What's Changed - test(responses): bump deprecated gemini-3-pro-preview to gemini-3.1-pro-preview by [@mateo-berri](https://github.com/mateo-berri) in [#29433](https://github.com/BerriAI/litellm/pull/29433) - fix: map mistral/ministral-8b-latest in model price map by [@mateo-berri](https://github.com/mateo-berri) in [#29453](https://github.com/BerriAI/litellm/pull/29453) - fix(datadog): split oversized batches on 413 instead of re-queueing forever by [@yassin-berriai](https://github.com/yassin-berriai) in [#29444](https://github.com/BerriAI/litellm/pull/29444) - feat(otel): allowlist team\_metadata sub-keys promoted to baggage by [@yassin-berriai](https://github.com/yassin-berriai) in [#29442](https://github.com/BerriAI/litellm/pull/29442) - fix: stop use\_chat\_completions\_api flag from leaking into provider request body by [@mateo-berri](https://github.com/mateo-berri) in [#29447](https://github.com/BerriAI/litellm/pull/29447) - fix(anthropic, fireworks): inline legacy $ref defs in tool schemas by [@milan-berri](https://github.com/milan-berri) in [#28646](https://github.com/BerriAI/litellm/pull/28646) - fix(proxy): omit OpenAI \[DONE] on google-genai streamGenerateContent by [@Sameerlite](https://github.com/Sameerlite) in [#29426](https://github.com/BerriAI/litellm/pull/29426) - ci(release): create stable/X.Y.x line branch on X.Y.0 tags by [@yuneng-berri](https://github.com/yuneng-berri) in [#29457](https://github.com/BerriAI/litellm/pull/29457) - fix(vector-stores): support engines URL for Vertex AI Search by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#27885](https://github.com/BerriAI/litellm/pull/27885) - fix(ui): render caller-supplied filter options in caller order by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29462](https://github.com/BerriAI/litellm/pull/29462) - fix(batches): skip unnecessary batch input file reads by [@Sameerlite](https://github.com/Sameerlite) in [#29114](https://github.com/BerriAI/litellm/pull/29114) - docs(agents): clarify when to create new test files by [@Sameerlite](https://github.com/Sameerlite) in [#29472](https://github.com/BerriAI/litellm/pull/29472) - Litellm OSS Staging by [@Sameerlite](https://github.com/Sameerlite) in [#29161](https://github.com/BerriAI/litellm/pull/29161) - fix(mcp): clear allowed\_tools and tool overrides on MCP server edit by [@Sameerlite](https://github.com/Sameerlite) in [#29411](https://github.com/BerriAI/litellm/pull/29411) - Litellm OSS Staging 010626 by [@Sameerlite](https://github.com/Sameerlite) in [#29422](https://github.com/BerriAI/litellm/pull/29422) - fix(ci): make CircleCI rerun-failed-tests collect tests when 2+ test files fail by [@mateo-berri](https://github.com/mateo-berri) in [#29475](https://github.com/BerriAI/litellm/pull/29475) - feat(a2a): watsonx Orchestrate agent provider by [@Sameerlite](https://github.com/Sameerlite) in [#29410](https://github.com/BerriAI/litellm/pull/29410) - fix(azure\_ai): strip tool-level extra fields on 400 and retry by [@Sameerlite](https://github.com/Sameerlite) in [#29479](https://github.com/BerriAI/litellm/pull/29479) - fix(docs): remove fixed dimensions from README hero image by [@mateo-berri](https://github.com/mateo-berri) in [#29496](https://github.com/BerriAI/litellm/pull/29496) - Litellm oss staging by [@Sameerlite](https://github.com/Sameerlite) in [#29492](https://github.com/BerriAI/litellm/pull/29492) - fix: small CLAUDE.md nits by [@mateo-berri](https://github.com/mateo-berri) in [#29504](https://github.com/BerriAI/litellm/pull/29504) - Add MCP semantic conventions to otelv2 by [@yassin-berriai](https://github.com/yassin-berriai) in [#29468](https://github.com/BerriAI/litellm/pull/29468) - fix(passthrough): emit otel guardrail span when a guardrail blocks by [@yassin-berriai](https://github.com/yassin-berriai) in [#29470](https://github.com/BerriAI/litellm/pull/29470) - fix(proxy): strip NUL bytes from spend log payloads to prevent PostgreSQL 22P05 by [@milan-berri](https://github.com/milan-berri) in [#29515](https://github.com/BerriAI/litellm/pull/29515) - \[internal copy of [#28008](https://github.com/BerriAI/litellm/issues/28008)] Support MCP OAuth passthrough and issuer-scoped JWT auth by [@mateo-berri](https://github.com/mateo-berri) in [#28356](https://github.com/BerriAI/litellm/pull/28356) - feat(vector-stores): forward per-request params to Vertex AI Search by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29459](https://github.com/BerriAI/litellm/pull/29459) - feat(proxy): add per-MCP-server RPM rate limiting for keys and teams by [@Sameerlite](https://github.com/Sameerlite) in [#29482](https://github.com/BerriAI/litellm/pull/29482) - fix(tests): drop module-level test calls that break local\_testing collection by [@mateo-berri](https://github.com/mateo-berri) in [#29520](https://github.com/BerriAI/litellm/pull/29520) - feat(agents): add LangFlow agent provider with A2A session bridging by [@Sameerlite](https://github.com/Sameerlite) in [#28963](https://github.com/BerriAI/litellm/pull/28963) - fix(ui/agents): make A2A skill tags enterable and validated by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29512](https://github.com/BerriAI/litellm/pull/29512) - \[internal copy of [#29232](https://github.com/BerriAI/litellm/issues/29232)] feat: route future Claude models to Anthropic provider via pattern matching by [@mateo-berri](https://github.com/mateo-berri) in [#29239](https://github.com/BerriAI/litellm/pull/29239) - fix(tests): drop import-time completion call in test\_register\_model by [@mateo-berri](https://github.com/mateo-berri) in [#29521](https://github.com/BerriAI/litellm/pull/29521) - test: stabilize batch VCR coverage and stop live upload/network leaks by [@mateo-berri](https://github.com/mateo-berri) in [#29477](https://github.com/BerriAI/litellm/pull/29477) - \[internal copy of [#29003](https://github.com/BerriAI/litellm/issues/29003)] fix(vertex\_ai): use user-supplied api\_base as is for Model Garden OpenAI-compat path by [@mateo-berri](https://github.com/mateo-berri) in [#29530](https://github.com/BerriAI/litellm/pull/29530) - feat(proxy): native /health/drain preStop hook for graceful shutdown by [@yassin-berriai](https://github.com/yassin-berriai) in [#29439](https://github.com/BerriAI/litellm/pull/29439) - fix(auth): preserve 401 status for expired JWTs in OTel traces by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29510](https://github.com/BerriAI/litellm/pull/29510) - fix(otel): capture 401 error details in management endpoint spans by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29535](https://github.com/BerriAI/litellm/pull/29535) - test(proxy/utils): pin bottom-of-file helper behavior by [@yuneng-berri](https://github.com/yuneng-berri) in [#29509](https://github.com/BerriAI/litellm/pull/29509) - test(proxy/utils): pin PrismaClient and spend-update behavior by [@yuneng-berri](https://github.com/yuneng-berri) in [#29488](https://github.com/BerriAI/litellm/pull/29488) - test(proxy/utils): pin ProxyLogging behavior by [@yuneng-berri](https://github.com/yuneng-berri) in [#29485](https://github.com/BerriAI/litellm/pull/29485) - fix: missing span for guardrail passthrough by [@yassin-berriai](https://github.com/yassin-berriai) in [#29552](https://github.com/BerriAI/litellm/pull/29552) - fix(auth): let internal users view search tools by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29542](https://github.com/BerriAI/litellm/pull/29542) - fix: missing mcp otel attributes by [@yassin-berriai](https://github.com/yassin-berriai) in [#29554](https://github.com/BerriAI/litellm/pull/29554) - fix(proxy): resolve managed video model ids for auth by [@shivamrawat1](https://github.com/shivamrawat1) in [#29545](https://github.com/BerriAI/litellm/pull/29545) - fix(key\_generate): allow team members to create keys on org-scoped teams by [@milan-berri](https://github.com/milan-berri) in [#29310](https://github.com/BerriAI/litellm/pull/29310) - test(pass-through): move Gemini pass-through tests to gemini-3.1-flash-lite by [@mateo-berri](https://github.com/mateo-berri) in [#29595](https://github.com/BerriAI/litellm/pull/29595) - Litellm oss staging 030626 by [@Sameerlite](https://github.com/Sameerlite) in [#29578](https://github.com/BerriAI/litellm/pull/29578) - Fix : a2a bugs 030626 by [@Sameerlite](https://github.com/Sameerlite) in [#29566](https://github.com/BerriAI/litellm/pull/29566) - \[internal copy of [#29533](https://github.com/BerriAI/litellm/issues/29533)] fix(anthropic/adapter): emit thinking block for reasoning\_content-only streaming chunks by [@mateo-berri](https://github.com/mateo-berri) in [#29600](https://github.com/BerriAI/litellm/pull/29600) - ci: reproduce default-Windows wheel install to guard MAX\_PATH by [@yuneng-berri](https://github.com/yuneng-berri) in [#29597](https://github.com/BerriAI/litellm/pull/29597) - fix(vertex): strip output\_config.effort for Vertex Claude models that reject it (Haiku 4.5) by [@mateo-berri](https://github.com/mateo-berri) in [#29585](https://github.com/BerriAI/litellm/pull/29585) - Litellm websocket improvements by [@Sameerlite](https://github.com/Sameerlite) in [#29563](https://github.com/BerriAI/litellm/pull/29563) - feat(arize/phoenix): OpenInference rendering parity — tool\_calls, cost, passthrough I/O, session/user, multimodal, cache tokens by [@milan-berri](https://github.com/milan-berri) in [#28800](https://github.com/BerriAI/litellm/pull/28800) - \[internal copy of [#29550](https://github.com/BerriAI/litellm/issues/29550)] fix: passthrough endpoints duplicate logs by [@mateo-berri](https://github.com/mateo-berri) in [#29598](https://github.com/BerriAI/litellm/pull/29598) - fix(ci): keep coverage rename green when a parallel node runs no tests by [@mateo-berri](https://github.com/mateo-berri) in [#29608](https://github.com/BerriAI/litellm/pull/29608) - test(vcr): close out the remaining VCR live-call leaks by [@mateo-berri](https://github.com/mateo-berri) in [#29603](https://github.com/BerriAI/litellm/pull/29603) - fix(key\_generate): exempt UI/CLI session tokens from the budget ceiling for team keys by [@yuneng-berri](https://github.com/yuneng-berri) in [#29612](https://github.com/BerriAI/litellm/pull/29612) - fix(realtime): allow null transcripts in stream logging payloads by [@milan-berri](https://github.com/milan-berri) in [#29625](https://github.com/BerriAI/litellm/pull/29625) - build(ui): migrate eslint to flat config + bump eslint-config-next to 16 by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29626](https://github.com/BerriAI/litellm/pull/29626) - fix(key\_generate): scope session-token team-key budget exemption to caller-supplied team\_id by [@yuneng-berri](https://github.com/yuneng-berri) in [#29641](https://github.com/BerriAI/litellm/pull/29641) - fix(proxy): disable proxy buffering on streaming SSE responses by [@mateo-berri](https://github.com/mateo-berri) in [#29557](https://github.com/BerriAI/litellm/pull/29557) - fix(mcp): gate /public/mcp\_hub strictly on litellm.public\_mcp\_servers by [@michelligabriele](https://github.com/michelligabriele) in [#27764](https://github.com/BerriAI/litellm/pull/27764) - ci(ui): frontend-lint job enforcing prettier + eslint on changed files by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29633](https://github.com/BerriAI/litellm/pull/29633) - fix(gemini): googleSearch + server-side tools and googleMaps JSON schema by [@Sameerlite](https://github.com/Sameerlite) in [#29582](https://github.com/BerriAI/litellm/pull/29582) - fix(proxy): passthrough 404 when SERVER\_ROOT\_PATH is set by [@Sameerlite](https://github.com/Sameerlite) in [#29658](https://github.com/BerriAI/litellm/pull/29658) - fix(gemini-realtime): use GA event names for Pipecat 1.3.x compatibility by [@Sameerlite](https://github.com/Sameerlite) in [#29662](https://github.com/BerriAI/litellm/pull/29662) - Litellm oss staging 040626 by [@Sameerlite](https://github.com/Sameerlite) in [#29671](https://github.com/BerriAI/litellm/pull/29671) - style(ui): prettier formatting pass over the dashboard by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29622](https://github.com/BerriAI/litellm/pull/29622) - chore: ignore prettier dashboard reformat in git blame by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29695](https://github.com/BerriAI/litellm/pull/29695) - fix(helm): Enable Backend Deployment to mount Gateway config.yaml by [@tin-berri](https://github.com/tin-berri) in [#29605](https://github.com/BerriAI/litellm/pull/29605) - \[internal copy of [#29277](https://github.com/BerriAI/litellm/issues/29277)] fix(proxy): add default=None to LiteLLM\_TeamMembership.litellm\_budget\_table by [@mateo-berri](https://github.com/mateo-berri) in [#29684](https://github.com/BerriAI/litellm/pull/29684) - test: make custom\_tokenizer proxy tests hermetic by [@yuneng-berri](https://github.com/yuneng-berri) in [#29643](https://github.com/BerriAI/litellm/pull/29643) - test(proxy): stop running real-DB tests in GitHub Actions unit jobs by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29700](https://github.com/BerriAI/litellm/pull/29700) - chore(ui): remove the bare-fetch lint rule by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29712](https://github.com/BerriAI/litellm/pull/29712) - Litellm jwt mapping virtualkeys by [@shivamrawat1](https://github.com/shivamrawat1) in [#28510](https://github.com/BerriAI/litellm/pull/28510) - refactor(ui): shared HTTP client + location-pinned fetch() lint rule by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29723](https://github.com/BerriAI/litellm/pull/29723) - fix(proxy): stop team BYOK model name corruption on model edit by [@yuneng-berri](https://github.com/yuneng-berri) in [#29731](https://github.com/BerriAI/litellm/pull/29731) - \[internal copy of [#29511](https://github.com/BerriAI/litellm/issues/29511)] feat(guardrails): add sensitive data routing to on-premise models by [@mateo-berri](https://github.com/mateo-berri) in [#29531](https://github.com/BerriAI/litellm/pull/29531) - fix(proxy/hooks): populate llm\_provider on internal rate-limit errors by [@mateo-berri](https://github.com/mateo-berri) in [#27707](https://github.com/BerriAI/litellm/pull/27707) - fix(vertex/anthropic): handle namespace tools and strip client\_metadata for codex compatibility by [@Sameerlite](https://github.com/Sameerlite) in [#29489](https://github.com/BerriAI/litellm/pull/29489) - Support OAuth M2M for Databricks Apps A2A agents by [@mateo-berri](https://github.com/mateo-berri) in [#29586](https://github.com/BerriAI/litellm/pull/29586) - fix: small CLAUDE.md nit by [@mateo-berri](https://github.com/mateo-berri) in [#29749](https://github.com/BerriAI/litellm/pull/29749) - fix(anthropic): route Claude Opus 4.8 through adaptive thinking by [@mateo-berri](https://github.com/mateo-berri) in [#29702](https://github.com/BerriAI/litellm/pull/29702) - fix(proxy): persist oauth2\_flow on MCP server registration by [@michelligabriele](https://github.com/michelligabriele) in [#29690](https://github.com/BerriAI/litellm/pull/29690) - \[internal copy of [#27491](https://github.com/BerriAI/litellm/issues/27491)] fix(realtime): Fix Realtime Audio Token Cost Tracking by [@mateo-berri](https://github.com/mateo-berri) in [#29722](https://github.com/BerriAI/litellm/pull/29722) - fix(galileo): use ingest traces API and standard logging payload by [@Sameerlite](https://github.com/Sameerlite) in [#29651](https://github.com/BerriAI/litellm/pull/29651) - fix(auth): expand all-team-models sentinel in can\_key\_call\_model for batch validation by [@Sameerlite](https://github.com/Sameerlite) in [#29746](https://github.com/BerriAI/litellm/pull/29746) - test(vcr): stop refreshing cassette TTL on read so cassettes lapse after 24h by [@mateo-berri](https://github.com/mateo-berri) in [#29784](https://github.com/BerriAI/litellm/pull/29784) - test(ci): record/replay OpenAI image gen so the spend E2E isn't outage-bound by [@mateo-berri](https://github.com/mateo-berri) in [#29787](https://github.com/BerriAI/litellm/pull/29787) - fix(ui): route MCP playground auth by oauth2 mode instead of token\_url by [@tin-berri](https://github.com/tin-berri) in [#29714](https://github.com/BerriAI/litellm/pull/29714) - refactor(ui): centralize proxy base URL resolution into tested resolver by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29793](https://github.com/BerriAI/litellm/pull/29793) - Litellm oss staging 050626 by [@Sameerlite](https://github.com/Sameerlite) in [#29774](https://github.com/BerriAI/litellm/pull/29774) - test(google): add google-genai SDK proxy integration tests by [@Sameerlite](https://github.com/Sameerlite) in [#29781](https://github.com/BerriAI/litellm/pull/29781) - fix(jwt): use resolved DB user\_id for spend on legacy email match by [@milan-berri](https://github.com/milan-berri) in [#29217](https://github.com/BerriAI/litellm/pull/29217) - feat(ui): generate dashboard API types from the proxy OpenAPI spec by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29816](https://github.com/BerriAI/litellm/pull/29816) - fix(proxy): drop deleted team BYOK model name from team.models by [@yuneng-berri](https://github.com/yuneng-berri) in [#29820](https://github.com/BerriAI/litellm/pull/29820) - feat(mcp): per-server env vars with global + per-user scopes by [@mateo-berri](https://github.com/mateo-berri) in [#28917](https://github.com/BerriAI/litellm/pull/28917) - refactor(ui): route behavior-preserving networking calls through apiClient by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29806](https://github.com/BerriAI/litellm/pull/29806) - fix(mcp): persist Tools-tab MCP OAuth token to DB by [@tin-berri](https://github.com/tin-berri) in [#29809](https://github.com/BerriAI/litellm/pull/29809) - fix(ui): require new expiration when regenerating an expired key by [@milan-berri](https://github.com/milan-berri) in [#29838](https://github.com/BerriAI/litellm/pull/29838) - refactor(ui): route query-building networking calls through apiClient by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29815](https://github.com/BerriAI/litellm/pull/29815) - Make the image-gen record/replay proxy report cache mode and per-request HIT/MISS by [@mateo-berri](https://github.com/mateo-berri) in [#29802](https://github.com/BerriAI/litellm/pull/29802) - feat(proxy): hot-reload .env in dev when running with --reload by [@mateo-berri](https://github.com/mateo-berri) in [#29783](https://github.com/BerriAI/litellm/pull/29783) - fix(ui): stop MCP playground tool calls from sending twice by [@tin-berri](https://github.com/tin-berri) in [#29821](https://github.com/BerriAI/litellm/pull/29821) - feat(fal\_ai): add Nano Banana / Gemini 2.5 Flash Image generation support by [@mateo-berri](https://github.com/mateo-berri) in [#29798](https://github.com/BerriAI/litellm/pull/29798) - Title: Fix managed batch cancel credential resolution by [@shivamrawat1](https://github.com/shivamrawat1) in [#29734](https://github.com/BerriAI/litellm/pull/29734) - Title: fix(proxy): resolve vector store file list credentials from team deployments by [@shivamrawat1](https://github.com/shivamrawat1) in [#29739](https://github.com/BerriAI/litellm/pull/29739) - refactor: convert AWS and GCP Terraform stacks into reusable modules … by [@yassin-berriai](https://github.com/yassin-berriai) in [#28103](https://github.com/BerriAI/litellm/pull/28103) - chore(ui): build ui for release by [@yuneng-berri](https://github.com/yuneng-berri) in [#29853](https://github.com/BerriAI/litellm/pull/29853) - fix(terraform/gcp): prompt for image\_registry in DeployStack one-click by [@yassin-berriai](https://github.com/yassin-berriai) in [#29852](https://github.com/BerriAI/litellm/pull/29852) - fix(terraform/gcp): abandon SQL user on destroy by [@yassin-berriai](https://github.com/yassin-berriai) in [#29855](https://github.com/BerriAI/litellm/pull/29855) - Extend the record/replay proxy to chat, embeddings, moderations, rerank, and Anthropic by [@mateo-berri](https://github.com/mateo-berri) in [#29847](https://github.com/BerriAI/litellm/pull/29847) - chore(deps): bump deps by [@yuneng-berri](https://github.com/yuneng-berri) in [#29860](https://github.com/BerriAI/litellm/pull/29860) - chore(ci): promote internal staging to main by [@yuneng-berri](https://github.com/yuneng-berri) in [#29861](https://github.com/BerriAI/litellm/pull/29861) - fix: 400 on Anthropic context overflow; seed identity on failed auth by [@yassin-berriai](https://github.com/yassin-berriai) in [#29848](https://github.com/BerriAI/litellm/pull/29848) - chore(ci): promote internal staging to main by [@yuneng-berri](https://github.com/yuneng-berri) in [#29862](https://github.com/BerriAI/litellm/pull/29862) - chore(release): patch v1.89.0-rc.1 with [#30064](https://github.com/BerriAI/litellm/issues/30064) (Claude Fable 5) for v1.89.0-rc.2 by [@mateo-berri](https://github.com/mateo-berri) in [#30143](https://github.com/BerriAI/litellm/pull/30143) **Full Changelog**: <https://github.com/BerriAI/litellm/compare/v1.88.0...v1.89.0> ### [`v1.89.0`](https://github.com/BerriAI/litellm/releases/tag/v1.89.0) [Compare Source](https://github.com/BerriAI/litellm/compare/v1.88.2...v1.89.0) ##### Verify Docker Image Signature All LiteLLM Docker images are signed with [cosign](https://docs.sigstore.dev/cosign/overview/). Every release is signed with the same key introduced in [commit `0112e53`](https://github.com/BerriAI/litellm/commit/0112e53046018d726492c814b3644b7d376029d0). **Verify using the pinned commit hash (recommended):** A commit hash is cryptographically immutable, so this is the strongest way to ensure you are using the original signing key: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/0112e53046018d726492c814b3644b7d376029d0/cosign.pub \ ghcr.io/berriai/litellm:v1.89.0 ``` **Verify using the release tag (convenience):** Tags are protected in this repository and resolve to the same key. This option is easier to read but relies on tag protection rules: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/v1.89.0/cosign.pub \ ghcr.io/berriai/litellm:v1.89.0 ``` Expected output: ``` The following checks were performed on each of these signatures: - The cosign claims were validated - The signatures were verified against the specified public key ``` *** ##### What's Changed - test(responses): bump deprecated gemini-3-pro-preview to gemini-3.1-pro-preview by [@mateo-berri](https://github.com/mateo-berri) in [#29433](https://github.com/BerriAI/litellm/pull/29433) - fix: map mistral/ministral-8b-latest in model price map by [@mateo-berri](https://github.com/mateo-berri) in [#29453](https://github.com/BerriAI/litellm/pull/29453) - fix(datadog): split oversized batches on 413 instead of re-queueing forever by [@yassin-berriai](https://github.com/yassin-berriai) in [#29444](https://github.com/BerriAI/litellm/pull/29444) - feat(otel): allowlist team\_metadata sub-keys promoted to baggage by [@yassin-berriai](https://github.com/yassin-berriai) in [#29442](https://github.com/BerriAI/litellm/pull/29442) - fix: stop use\_chat\_completions\_api flag from leaking into provider request body by [@mateo-berri](https://github.com/mateo-berri) in [#29447](https://github.com/BerriAI/litellm/pull/29447) - fix(anthropic, fireworks): inline legacy $ref defs in tool schemas by [@milan-berri](https://github.com/milan-berri) in [#28646](https://github.com/BerriAI/litellm/pull/28646) - fix(proxy): omit OpenAI \[DONE] on google-genai streamGenerateContent by [@Sameerlite](https://github.com/Sameerlite) in [#29426](https://github.com/BerriAI/litellm/pull/29426) - ci(release): create stable/X.Y.x line branch on X.Y.0 tags by [@yuneng-berri](https://github.com/yuneng-berri) in [#29457](https://github.com/BerriAI/litellm/pull/29457) - fix(vector-stores): support engines URL for Vertex AI Search by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#27885](https://github.com/BerriAI/litellm/pull/27885) - fix(ui): render caller-supplied filter options in caller order by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29462](https://github.com/BerriAI/litellm/pull/29462) - fix(batches): skip unnecessary batch input file reads by [@Sameerlite](https://github.com/Sameerlite) in [#29114](https://github.com/BerriAI/litellm/pull/29114) - docs(agents): clarify when to create new test files by [@Sameerlite](https://github.com/Sameerlite) in [#29472](https://github.com/BerriAI/litellm/pull/29472) - Litellm OSS Staging by [@Sameerlite](https://github.com/Sameerlite) in [#29161](https://github.com/BerriAI/litellm/pull/29161) - fix(mcp): clear allowed\_tools and tool overrides on MCP server edit by [@Sameerlite](https://github.com/Sameerlite) in [#29411](https://github.com/BerriAI/litellm/pull/29411) - Litellm OSS Staging 010626 by [@Sameerlite](https://github.com/Sameerlite) in [#29422](https://github.com/BerriAI/litellm/pull/29422) - fix(ci): make CircleCI rerun-failed-tests collect tests when 2+ test files fail by [@mateo-berri](https://github.com/mateo-berri) in [#29475](https://github.com/BerriAI/litellm/pull/29475) - feat(a2a): watsonx Orchestrate agent provider by [@Sameerlite](https://github.com/Sameerlite) in [#29410](https://github.com/BerriAI/litellm/pull/29410) - fix(azure\_ai): strip tool-level extra fields on 400 and retry by [@Sameerlite](https://github.com/Sameerlite) in [#29479](https://github.com/BerriAI/litellm/pull/29479) - fix(docs): remove fixed dimensions from README hero image by [@mateo-berri](https://github.com/mateo-berri) in [#29496](https://github.com/BerriAI/litellm/pull/29496) - Litellm oss staging by [@Sameerlite](https://github.com/Sameerlite) in [#29492](https://github.com/BerriAI/litellm/pull/29492) - fix: small CLAUDE.md nits by [@mateo-berri](https://github.com/mateo-berri) in [#29504](https://github.com/BerriAI/litellm/pull/29504) - Add MCP semantic conventions to otelv2 by [@yassin-berriai](https://github.com/yassin-berriai) in [#29468](https://github.com/BerriAI/litellm/pull/29468) - fix(passthrough): emit otel guardrail span when a guardrail blocks by [@yassin-berriai](https://github.com/yassin-berriai) in [#29470](https://github.com/BerriAI/litellm/pull/29470) - fix(proxy): strip NUL bytes from spend log payloads to prevent PostgreSQL 22P05 by [@milan-berri](https://github.com/milan-berri) in [#29515](https://github.com/BerriAI/litellm/pull/29515) - \[internal copy of [#28008](https://github.com/BerriAI/litellm/issues/28008)] Support MCP OAuth passthrough and issuer-scoped JWT auth by [@mateo-berri](https://github.com/mateo-berri) in [#28356](https://github.com/BerriAI/litellm/pull/28356) - feat(vector-stores): forward per-request params to Vertex AI Search by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29459](https://github.com/BerriAI/litellm/pull/29459) - feat(proxy): add per-MCP-server RPM rate limiting for keys and teams by [@Sameerlite](https://github.com/Sameerlite) in [#29482](https://github.com/BerriAI/litellm/pull/29482) - fix(tests): drop module-level test calls that break local\_testing collection by [@mateo-berri](https://github.com/mateo-berri) in [#29520](https://github.com/BerriAI/litellm/pull/29520) - feat(agents): add LangFlow agent provider with A2A session bridging by [@Sameerlite](https://github.com/Sameerlite) in [#28963](https://github.com/BerriAI/litellm/pull/28963) - fix(ui/agents): make A2A skill tags enterable and validated by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29512](https://github.com/BerriAI/litellm/pull/29512) - \[internal copy of [#29232](https://github.com/BerriAI/litellm/issues/29232)] feat: route future Claude models to Anthropic provider via pattern matching by [@mateo-berri](https://github.com/mateo-berri) in [#29239](https://github.com/BerriAI/litellm/pull/29239) - fix(tests): drop import-time completion call in test\_register\_model by [@mateo-berri](https://github.com/mateo-berri) in [#29521](https://github.com/BerriAI/litellm/pull/29521) - test: stabilize batch VCR coverage and stop live upload/network leaks by [@mateo-berri](https://github.com/mateo-berri) in [#29477](https://github.com/BerriAI/litellm/pull/29477) - \[internal copy of [#29003](https://github.com/BerriAI/litellm/issues/29003)] fix(vertex\_ai): use user-supplied api\_base as is for Model Garden OpenAI-compat path by [@mateo-berri](https://github.com/mateo-berri) in [#29530](https://github.com/BerriAI/litellm/pull/29530) - feat(proxy): native /health/drain preStop hook for graceful shutdown by [@yassin-berriai](https://github.com/yassin-berriai) in [#29439](https://github.com/BerriAI/litellm/pull/29439) - fix(auth): preserve 401 status for expired JWTs in OTel traces by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29510](https://github.com/BerriAI/litellm/pull/29510) - fix(otel): capture 401 error details in management endpoint spans by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29535](https://github.com/BerriAI/litellm/pull/29535) - test(proxy/utils): pin bottom-of-file helper behavior by [@yuneng-berri](https://github.com/yuneng-berri) in [#29509](https://github.com/BerriAI/litellm/pull/29509) - test(proxy/utils): pin PrismaClient and spend-update behavior by [@yuneng-berri](https://github.com/yuneng-berri) in [#29488](https://github.com/BerriAI/litellm/pull/29488) - test(proxy/utils): pin ProxyLogging behavior by [@yuneng-berri](https://github.com/yuneng-berri) in [#29485](https://github.com/BerriAI/litellm/pull/29485) - fix: missing span for guardrail passthrough by [@yassin-berriai](https://github.com/yassin-berriai) in [#29552](https://github.com/BerriAI/litellm/pull/29552) - fix(auth): let internal users view search tools by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29542](https://github.com/BerriAI/litellm/pull/29542) - fix: missing mcp otel attributes by [@yassin-berriai](https://github.com/yassin-berriai) in [#29554](https://github.com/BerriAI/litellm/pull/29554) - fix(proxy): resolve managed video model ids for auth by [@shivamrawat1](https://github.com/shivamrawat1) in [#29545](https://github.com/BerriAI/litellm/pull/29545) - fix(key\_generate): allow team members to create keys on org-scoped teams by [@milan-berri](https://github.com/milan-berri) in [#29310](https://github.com/BerriAI/litellm/pull/29310) - test(pass-through): move Gemini pass-through tests to gemini-3.1-flash-lite by [@mateo-berri](https://github.com/mateo-berri) in [#29595](https://github.com/BerriAI/litellm/pull/29595) - Litellm oss staging 030626 by [@Sameerlite](https://github.com/Sameerlite) in [#29578](https://github.com/BerriAI/litellm/pull/29578) - Fix : a2a bugs 030626 by [@Sameerlite](https://github.com/Sameerlite) in [#29566](https://github.com/BerriAI/litellm/pull/29566) - \[internal copy of [#29533](https://github.com/BerriAI/litellm/issues/29533)] fix(anthropic/adapter): emit thinking block for reasoning\_content-only streaming chunks by [@mateo-berri](https://github.com/mateo-berri) in [#29600](https://github.com/BerriAI/litellm/pull/29600) - ci: reproduce default-Windows wheel install to guard MAX\_PATH by [@yuneng-berri](https://github.com/yuneng-berri) in [#29597](https://github.com/BerriAI/litellm/pull/29597) - fix(vertex): strip output\_config.effort for Vertex Claude models that reject it (Haiku 4.5) by [@mateo-berri](https://github.com/mateo-berri) in [#29585](https://github.com/BerriAI/litellm/pull/29585) - Litellm websocket improvements by [@Sameerlite](https://github.com/Sameerlite) in [#29563](https://github.com/BerriAI/litellm/pull/29563) - feat(arize/phoenix): OpenInference rendering parity — tool\_calls, cost, passthrough I/O, session/user, multimodal, cache tokens by [@milan-berri](https://github.com/milan-berri) in [#28800](https://github.com/BerriAI/litellm/pull/28800) - \[internal copy of [#29550](https://github.com/BerriAI/litellm/issues/29550)] fix: passthrough endpoints duplicate logs by [@mateo-berri](https://github.com/mateo-berri) in [#29598](https://github.com/BerriAI/litellm/pull/29598) - fix(ci): keep coverage rename green when a parallel node runs no tests by [@mateo-berri](https://github.com/mateo-berri) in [#29608](https://github.com/BerriAI/litellm/pull/29608) - test(vcr): close out the remaining VCR live-call leaks by [@mateo-berri](https://github.com/mateo-berri) in [#29603](https://github.com/BerriAI/litellm/pull/29603) - fix(key\_generate): exempt UI/CLI session tokens from the budget ceiling for team keys by [@yuneng-berri](https://github.com/yuneng-berri) in [#29612](https://github.com/BerriAI/litellm/pull/29612) - fix(realtime): allow null transcripts in stream logging payloads by [@milan-berri](https://github.com/milan-berri) in [#29625](https://github.com/BerriAI/litellm/pull/29625) - build(ui): migrate eslint to flat config + bump eslint-config-next to 16 by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29626](https://github.com/BerriAI/litellm/pull/29626) - fix(key\_generate): scope session-token team-key budget exemption to caller-supplied team\_id by [@yuneng-berri](https://github.com/yuneng-berri) in [#29641](https://github.com/BerriAI/litellm/pull/29641) - fix(proxy): disable proxy buffering on streaming SSE responses by [@mateo-berri](https://github.com/mateo-berri) in [#29557](https://github.com/BerriAI/litellm/pull/29557) - fix(mcp): gate /public/mcp\_hub strictly on litellm.public\_mcp\_servers by [@michelligabriele](https://github.com/michelligabriele) in [#27764](https://github.com/BerriAI/litellm/pull/27764) - ci(ui): frontend-lint job enforcing prettier + eslint on changed files by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29633](https://github.com/BerriAI/litellm/pull/29633) - fix(gemini): googleSearch + server-side tools and googleMaps JSON schema by [@Sameerlite](https://github.com/Sameerlite) in [#29582](https://github.com/BerriAI/litellm/pull/29582) - fix(proxy): passthrough 404 when SERVER\_ROOT\_PATH is set by [@Sameerlite](https://github.com/Sameerlite) in [#29658](https://github.com/BerriAI/litellm/pull/29658) - fix(gemini-realtime): use GA event names for Pipecat 1.3.x compatibility by [@Sameerlite](https://github.com/Sameerlite) in [#29662](https://github.com/BerriAI/litellm/pull/29662) - Litellm oss staging 040626 by [@Sameerlite](https://github.com/Sameerlite) in [#29671](https://github.com/BerriAI/litellm/pull/29671) - style(ui): prettier formatting pass over the dashboard by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29622](https://github.com/BerriAI/litellm/pull/29622) - chore: ignore prettier dashboard reformat in git blame by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29695](https://github.com/BerriAI/litellm/pull/29695) - fix(helm): Enable Backend Deployment to mount Gateway config.yaml by [@tin-berri](https://github.com/tin-berri) in [#29605](https://github.com/BerriAI/litellm/pull/29605) - \[internal copy of [#29277](https://github.com/BerriAI/litellm/issues/29277)] fix(proxy): add default=None to LiteLLM\_TeamMembership.litellm\_budget\_table by [@mateo-berri](https://github.com/mateo-berri) in [#29684](https://github.com/BerriAI/litellm/pull/29684) - test: make custom\_tokenizer proxy tests hermetic by [@yuneng-berri](https://github.com/yuneng-berri) in [#29643](https://github.com/BerriAI/litellm/pull/29643) - test(proxy): stop running real-DB tests in GitHub Actions unit jobs by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29700](https://github.com/BerriAI/litellm/pull/29700) - chore(ui): remove the bare-fetch lint rule by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29712](https://github.com/BerriAI/litellm/pull/29712) - Litellm jwt mapping virtualkeys by [@shivamrawat1](https://github.com/shivamrawat1) in [#28510](https://github.com/BerriAI/litellm/pull/28510) - refactor(ui): shared HTTP client + location-pinned fetch() lint rule by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29723](https://github.com/BerriAI/litellm/pull/29723) - fix(proxy): stop team BYOK model name corruption on model edit by [@yuneng-berri](https://github.com/yuneng-berri) in [#29731](https://github.com/BerriAI/litellm/pull/29731) - \[internal copy of [#29511](https://github.com/BerriAI/litellm/issues/29511)] feat(guardrails): add sensitive data routing to on-premise models by [@mateo-berri](https://github.com/mateo-berri) in [#29531](https://github.com/BerriAI/litellm/pull/29531) - fix(proxy/hooks): populate llm\_provider on internal rate-limit errors by [@mateo-berri](https://github.com/mateo-berri) in [#27707](https://github.com/BerriAI/litellm/pull/27707) - fix(vertex/anthropic): handle namespace tools and strip client\_metadata for codex compatibility by [@Sameerlite](https://github.com/Sameerlite) in [#29489](https://github.com/BerriAI/litellm/pull/29489) - Support OAuth M2M for Databricks Apps A2A agents by [@mateo-berri](https://github.com/mateo-berri) in [#29586](https://github.com/BerriAI/litellm/pull/29586) - fix: small CLAUDE.md nit by [@mateo-berri](https://github.com/mateo-berri) in [#29749](https://github.com/BerriAI/litellm/pull/29749) - fix(anthropic): route Claude Opus 4.8 through adaptive thinking by [@mateo-berri](https://github.com/mateo-berri) in [#29702](https://github.com/BerriAI/litellm/pull/29702) - fix(proxy): persist oauth2\_flow on MCP server registration by [@michelligabriele](https://github.com/michelligabriele) in [#29690](https://github.com/BerriAI/litellm/pull/29690) - \[internal copy of [#27491](https://github.com/BerriAI/litellm/issues/27491)] fix(realtime): Fix Realtime Audio Token Cost Tracking by [@mateo-berri](https://github.com/mateo-berri) in [#29722](https://github.com/BerriAI/litellm/pull/29722) - fix(galileo): use ingest traces API and standard logging payload by [@Sameerlite](https://github.com/Sameerlite) in [#29651](https://github.com/BerriAI/litellm/pull/29651) - fix(auth): expand all-team-models sentinel in can\_key\_call\_model for batch validation by [@Sameerlite](https://github.com/Sameerlite) in [#29746](https://github.com/BerriAI/litellm/pull/29746) - test(vcr): stop refreshing cassette TTL on read so cassettes lapse after 24h by [@mateo-berri](https://github.com/mateo-berri) in [#29784](https://github.com/BerriAI/litellm/pull/29784) - test(ci): record/replay OpenAI image gen so the spend E2E isn't outage-bound by [@mateo-berri](https://github.com/mateo-berri) in [#29787](https://github.com/BerriAI/litellm/pull/29787) - fix(ui): route MCP playground auth by oauth2 mode instead of token\_url by [@tin-berri](https://github.com/tin-berri) in [#29714](https://github.com/BerriAI/litellm/pull/29714) - refactor(ui): centralize proxy base URL resolution into tested resolver by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29793](https://github.com/BerriAI/litellm/pull/29793) - Litellm oss staging 050626 by [@Sameerlite](https://github.com/Sameerlite) in [#29774](https://github.com/BerriAI/litellm/pull/29774) - test(google): add google-genai SDK proxy integration tests by [@Sameerlite](https://github.com/Sameerlite) in [#29781](https://github.com/BerriAI/litellm/pull/29781) - fix(jwt): use resolved DB user\_id for spend on legacy email match by [@milan-berri](https://github.com/milan-berri) in [#29217](https://github.com/BerriAI/litellm/pull/29217) - feat(ui): generate dashboard API types from the proxy OpenAPI spec by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29816](https://github.com/BerriAI/litellm/pull/29816) - fix(proxy): drop deleted team BYOK model name from team.models by [@yuneng-berri](https://github.com/yuneng-berri) in [#29820](https://github.com/BerriAI/litellm/pull/29820) - feat(mcp): per-server env vars with global + per-user scopes by [@mateo-berri](https://github.com/mateo-berri) in [#28917](https://github.com/BerriAI/litellm/pull/28917) - refactor(ui): route behavior-preserving networking calls through apiClient by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29806](https://github.com/BerriAI/litellm/pull/29806) - fix(mcp): persist Tools-tab MCP OAuth token to DB by [@tin-berri](https://github.com/tin-berri) in [#29809](https://github.com/BerriAI/litellm/pull/29809) - fix(ui): require new expiration when regenerating an expired key by [@milan-berri](https://github.com/milan-berri) in [#29838](https://github.com/BerriAI/litellm/pull/29838) - refactor(ui): route query-building networking calls through apiClient by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29815](https://github.com/BerriAI/litellm/pull/29815) - Make the image-gen record/replay proxy report cache mode and per-request HIT/MISS by [@mateo-berri](https://github.com/mateo-berri) in [#29802](https://github.com/BerriAI/litellm/pull/29802) - feat(proxy): hot-reload .env in dev when running with --reload by [@mateo-berri](https://github.com/mateo-berri) in [#29783](https://github.com/BerriAI/litellm/pull/29783) - fix(ui): stop MCP playground tool calls from sending twice by [@tin-berri](https://github.com/tin-berri) in [#29821](https://github.com/BerriAI/litellm/pull/29821) - feat(fal\_ai): add Nano Banana / Gemini 2.5 Flash Image generation support by [@mateo-berri](https://github.com/mateo-berri) in [#29798](https://github.com/BerriAI/litellm/pull/29798) - Title: Fix managed batch cancel credential resolution by [@shivamrawat1](https://github.com/shivamrawat1) in [#29734](https://github.com/BerriAI/litellm/pull/29734) - Title: fix(proxy): resolve vector store file list credentials from team deployments by [@shivamrawat1](https://github.com/shivamrawat1) in [#29739](https://github.com/BerriAI/litellm/pull/29739) - refactor: convert AWS and GCP Terraform stacks into reusable modules … by [@yassin-berriai](https://github.com/yassin-berriai) in [#28103](https://github.com/BerriAI/litellm/pull/28103) - chore(ui): build ui for release by [@yuneng-berri](https://github.com/yuneng-berri) in [#29853](https://github.com/BerriAI/litellm/pull/29853) - fix(terraform/gcp): prompt for image\_registry in DeployStack one-click by [@yassin-berriai](https://github.com/yassin-berriai) in [#29852](https://github.com/BerriAI/litellm/pull/29852) - fix(terraform/gcp): abandon SQL user on destroy by [@yassin-berriai](https://github.com/yassin-berriai) in [#29855](https://github.com/BerriAI/litellm/pull/29855) - Extend the record/replay proxy to chat, embeddings, moderations, rerank, and Anthropic by [@mateo-berri](https://github.com/mateo-berri) in [#29847](https://github.com/BerriAI/litellm/pull/29847) - chore(deps): bump deps by [@yuneng-berri](https://github.com/yuneng-berri) in [#29860](https://github.com/BerriAI/litellm/pull/29860) - chore(ci): promote internal staging to main by [@yuneng-berri](https://github.com/yuneng-berri) in [#29861](https://github.com/BerriAI/litellm/pull/29861) - fix: 400 on Anthropic context overflow; seed identity on failed auth by [@yassin-berriai](https://github.com/yassin-berriai) in [#29848](https://github.com/BerriAI/litellm/pull/29848) - chore(ci): promote internal staging to main by [@yuneng-berri](https://github.com/yuneng-berri) in [#29862](https://github.com/BerriAI/litellm/pull/29862) - chore(release): patch v1.89.0-rc.1 with [#30064](https://github.com/BerriAI/litellm/issues/30064) (Claude Fable 5) for v1.89.0-rc.2 by [@mateo-berri](https://github.com/mateo-berri) in [#30143](https://github.com/BerriAI/litellm/pull/30143) **Full Changelog**: <https://github.com/BerriAI/litellm/compare/v1.88.0...v1.89.0> ### [`v1.88.2`](https://github.com/BerriAI/litellm/releases/tag/v1.88.2) [Compare Source](https://github.com/BerriAI/litellm/compare/v1.88.2...v1.88.2) ##### Verify Docker Image Signature All LiteLLM Docker images are signed with [cosign](https://docs.sigstore.dev/cosign/overview/). Every release is signed with the same key introduced in [commit `0112e53`](https://github.com/BerriAI/litellm/commit/0112e53046018d726492c814b3644b7d376029d0). **Verify using the pinned commit hash (recommended):** A commit hash is cryptographically immutable, so this is the strongest way to ensure you are using the original signing key: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/0112e53046018d726492c814b3644b7d376029d0/cosign.pub \ ghcr.io/berriai/litellm:v1.88.2 ``` **Verify using the release tag (convenience):** Tags are protected in this repository and resolve to the same key. This option is easier to read but relies on tag protection rules: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/v1.88.2/cosign.pub \ ghcr.io/berriai/litellm:v1.88.2 ``` Expected output: ``` The following checks were performed on each of these signatures: - The cosign claims were validated - The signatures were verified against the specified public key ``` *** ##### What's Changed - chore(release): backport Fable 5, batch-file auth, CrowdStrike AIDR, Mantle Responses SigV4, and NetApp streaming-cost fix to stable/1.88.x and cut 1.88.2 by [@mateo-berri](https://github.com/mateo-berri) in [#30144](https://github.com/BerriAI/litellm/pull/30144) - chore(release): backport DB-resilience, passthrough, model-info, budget, and deps fixes to stable/1.88.x by [@yuneng-berri](https://github.com/yuneng-berri) in [#30408](https://github.com/BerriAI/litellm/pull/30408) **Full Changelog**: <https://github.com/BerriAI/litellm/compare/v1.88.1...v1.88.2> ### [`v1.88.2`](https://github.com/BerriAI/litellm/releases/tag/v1.88.2) [Compare Source](https://github.com/BerriAI/litellm/compare/v1.88.1...v1.88.2) ##### Verify Docker Image Signature All LiteLLM Docker images are signed with [cosign](https://docs.sigstore.dev/cosign/overview/). Every release is signed with the same key introduced in [commit `0112e53`](https://github.com/BerriAI/litellm/commit/0112e53046018d726492c814b3644b7d376029d0). **Verify using the pinned commit hash (recommended):** A commit hash is cryptographically immutable, so this is the strongest way to ensure you are using the original signing key: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/0112e53046018d726492c814b3644b7d376029d0/cosign.pub \ ghcr.io/berriai/litellm:v1.88.2 ``` **Verify using the release tag (convenience):** Tags are protected in this repository and resolve to the same key. This option is easier to read but relies on tag protection rules: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/v1.88.2/cosign.pub \ ghcr.io/berriai/litellm:v1.88.2 ``` Expected output: ``` The following checks were performed on each of these signatures: - The cosign claims were validated - The signatures were verified against the specified public key ``` *** ##### What's Changed - chore(release): backport Fable 5, batch-file auth, CrowdStrike AIDR, Mantle Responses SigV4, and NetApp streaming-cost fix to stable/1.88.x and cut 1.88.2 by [@mateo-berri](https://github.com/mateo-berri) in [#30144](https://github.com/BerriAI/litellm/pull/30144) - chore(release): backport DB-resilience, passthrough, model-info, budget, and deps fixes to stable/1.88.x by [@yuneng-berri](https://github.com/yuneng-berri) in [#30408](https://github.com/BerriAI/litellm/pull/30408) **Full Changelog**: <https://github.com/BerriAI/litellm/compare/v1.88.1...v1.88.2> </details> --- ### Configuration 📅 **Schedule**: (in timezone Europe/London) - Branch creation - At any time (no schedule defined) - Automerge - At any time (no schedule defined) 🚦 **Automerge**: Disabled by config. Please merge this manually once you are satisfied. ♻ **Rebasing**: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox. 🔕 **Ignore**: Close this PR and you won't be reminded about these updates again. --- - [ ] If you want to rebase/retry this PR, check this box --- This PR has been generated by [Mend Renovate](https://github.com/renovatebot/renovate).  Reviewed-on: https://forgejo.hayden.moe/hayden/phoebe/pulls/93

…to v1.89.0 (#200) This PR contains the following updates: | Package | Update | Change | |---|---|---| | [https://github.com/BerriAI/litellm.git](https://github.com/BerriAI/litellm) | minor | `v1.85.1` → `v1.89.0` | --- > ⚠️ **Warning** > > Some dependencies could not be looked up. Check the [Dependency Dashboard](issues/155) for more information. --- ### Release Notes <details> <summary>BerriAI/litellm (https://github.com/BerriAI/litellm.git)</summary> ### [`v1.89.0`](https://github.com/BerriAI/litellm/releases/tag/v1.89.0) [Compare Source](https://github.com/BerriAI/litellm/compare/v1.88.2...v1.89.0) #### Verify Docker Image Signature All LiteLLM Docker images are signed with [cosign](https://docs.sigstore.dev/cosign/overview/). Every release is signed with the same key introduced in [commit `0112e53`](https://github.com/BerriAI/litellm/commit/0112e53046018d726492c814b3644b7d376029d0). **Verify using the pinned commit hash (recommended):** A commit hash is cryptographically immutable, so this is the strongest way to ensure you are using the original signing key: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/0112e53046018d726492c814b3644b7d376029d0/cosign.pub \ ghcr.io/berriai/litellm:v1.89.0 ``` **Verify using the release tag (convenience):** Tags are protected in this repository and resolve to the same key. This option is easier to read but relies on tag protection rules: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/v1.89.0/cosign.pub \ ghcr.io/berriai/litellm:v1.89.0 ``` Expected output: ``` The following checks were performed on each of these signatures: - The cosign claims were validated - The signatures were verified against the specified public key ``` *** #### What's Changed - test(responses): bump deprecated gemini-3-pro-preview to gemini-3.1-pro-preview by [@mateo-berri](https://github.com/mateo-berri) in [#29433](https://github.com/BerriAI/litellm/pull/29433) - fix: map mistral/ministral-8b-latest in model price map by [@mateo-berri](https://github.com/mateo-berri) in [#29453](https://github.com/BerriAI/litellm/pull/29453) - fix(datadog): split oversized batches on 413 instead of re-queueing forever by [@yassin-berriai](https://github.com/yassin-berriai) in [#29444](https://github.com/BerriAI/litellm/pull/29444) - feat(otel): allowlist team\_metadata sub-keys promoted to baggage by [@yassin-berriai](https://github.com/yassin-berriai) in [#29442](https://github.com/BerriAI/litellm/pull/29442) - fix: stop use\_chat\_completions\_api flag from leaking into provider request body by [@mateo-berri](https://github.com/mateo-berri) in [#29447](https://github.com/BerriAI/litellm/pull/29447) - fix(anthropic, fireworks): inline legacy $ref defs in tool schemas by [@milan-berri](https://github.com/milan-berri) in [#28646](https://github.com/BerriAI/litellm/pull/28646) - fix(proxy): omit OpenAI \[DONE] on google-genai streamGenerateContent by [@Sameerlite](https://github.com/Sameerlite) in [#29426](https://github.com/BerriAI/litellm/pull/29426) - ci(release): create stable/X.Y.x line branch on X.Y.0 tags by [@yuneng-berri](https://github.com/yuneng-berri) in [#29457](https://github.com/BerriAI/litellm/pull/29457) - fix(vector-stores): support engines URL for Vertex AI Search by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#27885](https://github.com/BerriAI/litellm/pull/27885) - fix(ui): render caller-supplied filter options in caller order by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29462](https://github.com/BerriAI/litellm/pull/29462) - fix(batches): skip unnecessary batch input file reads by [@Sameerlite](https://github.com/Sameerlite) in [#29114](https://github.com/BerriAI/litellm/pull/29114) - docs(agents): clarify when to create new test files by [@Sameerlite](https://github.com/Sameerlite) in [#29472](https://github.com/BerriAI/litellm/pull/29472) - Litellm OSS Staging by [@Sameerlite](https://github.com/Sameerlite) in [#29161](https://github.com/BerriAI/litellm/pull/29161) - fix(mcp): clear allowed\_tools and tool overrides on MCP server edit by [@Sameerlite](https://github.com/Sameerlite) in [#29411](https://github.com/BerriAI/litellm/pull/29411) - Litellm OSS Staging 010626 by [@Sameerlite](https://github.com/Sameerlite) in [#29422](https://github.com/BerriAI/litellm/pull/29422) - fix(ci): make CircleCI rerun-failed-tests collect tests when 2+ test files fail by [@mateo-berri](https://github.com/mateo-berri) in [#29475](https://github.com/BerriAI/litellm/pull/29475) - feat(a2a): watsonx Orchestrate agent provider by [@Sameerlite](https://github.com/Sameerlite) in [#29410](https://github.com/BerriAI/litellm/pull/29410) - fix(azure\_ai): strip tool-level extra fields on 400 and retry by [@Sameerlite](https://github.com/Sameerlite) in [#29479](https://github.com/BerriAI/litellm/pull/29479) - fix(docs): remove fixed dimensions from README hero image by [@mateo-berri](https://github.com/mateo-berri) in [#29496](https://github.com/BerriAI/litellm/pull/29496) - Litellm oss staging by [@Sameerlite](https://github.com/Sameerlite) in [#29492](https://github.com/BerriAI/litellm/pull/29492) - fix: small CLAUDE.md nits by [@mateo-berri](https://github.com/mateo-berri) in [#29504](https://github.com/BerriAI/litellm/pull/29504) - Add MCP semantic conventions to otelv2 by [@yassin-berriai](https://github.com/yassin-berriai) in [#29468](https://github.com/BerriAI/litellm/pull/29468) - fix(passthrough): emit otel guardrail span when a guardrail blocks by [@yassin-berriai](https://github.com/yassin-berriai) in [#29470](https://github.com/BerriAI/litellm/pull/29470) - fix(proxy): strip NUL bytes from spend log payloads to prevent PostgreSQL 22P05 by [@milan-berri](https://github.com/milan-berri) in [#29515](https://github.com/BerriAI/litellm/pull/29515) - \[internal copy of [#28008](https://github.com/BerriAI/litellm/issues/28008)] Support MCP OAuth passthrough and issuer-scoped JWT auth by [@mateo-berri](https://github.com/mateo-berri) in [#28356](https://github.com/BerriAI/litellm/pull/28356) - feat(vector-stores): forward per-request params to Vertex AI Search by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29459](https://github.com/BerriAI/litellm/pull/29459) - feat(proxy): add per-MCP-server RPM rate limiting for keys and teams by [@Sameerlite](https://github.com/Sameerlite) in [#29482](https://github.com/BerriAI/litellm/pull/29482) - fix(tests): drop module-level test calls that break local\_testing collection by [@mateo-berri](https://github.com/mateo-berri) in [#29520](https://github.com/BerriAI/litellm/pull/29520) - feat(agents): add LangFlow agent provider with A2A session bridging by [@Sameerlite](https://github.com/Sameerlite) in [#28963](https://github.com/BerriAI/litellm/pull/28963) - fix(ui/agents): make A2A skill tags enterable and validated by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29512](https://github.com/BerriAI/litellm/pull/29512) - \[internal copy of [#29232](https://github.com/BerriAI/litellm/issues/29232)] feat: route future Claude models to Anthropic provider via pattern matching by [@mateo-berri](https://github.com/mateo-berri) in [#29239](https://github.com/BerriAI/litellm/pull/29239) - fix(tests): drop import-time completion call in test\_register\_model by [@mateo-berri](https://github.com/mateo-berri) in [#29521](https://github.com/BerriAI/litellm/pull/29521) - test: stabilize batch VCR coverage and stop live upload/network leaks by [@mateo-berri](https://github.com/mateo-berri) in [#29477](https://github.com/BerriAI/litellm/pull/29477) - \[internal copy of [#29003](https://github.com/BerriAI/litellm/issues/29003)] fix(vertex\_ai): use user-supplied api\_base as is for Model Garden OpenAI-compat path by [@mateo-berri](https://github.com/mateo-berri) in [#29530](https://github.com/BerriAI/litellm/pull/29530) - feat(proxy): native /health/drain preStop hook for graceful shutdown by [@yassin-berriai](https://github.com/yassin-berriai) in [#29439](https://github.com/BerriAI/litellm/pull/29439) - fix(auth): preserve 401 status for expired JWTs in OTel traces by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29510](https://github.com/BerriAI/litellm/pull/29510) - fix(otel): capture 401 error details in management endpoint spans by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29535](https://github.com/BerriAI/litellm/pull/29535) - test(proxy/utils): pin bottom-of-file helper behavior by [@yuneng-berri](https://github.com/yuneng-berri) in [#29509](https://github.com/BerriAI/litellm/pull/29509) - test(proxy/utils): pin PrismaClient and spend-update behavior by [@yuneng-berri](https://github.com/yuneng-berri) in [#29488](https://github.com/BerriAI/litellm/pull/29488) - test(proxy/utils): pin ProxyLogging behavior by [@yuneng-berri](https://github.com/yuneng-berri) in [#29485](https://github.com/BerriAI/litellm/pull/29485) - fix: missing span for guardrail passthrough by [@yassin-berriai](https://github.com/yassin-berriai) in [#29552](https://github.com/BerriAI/litellm/pull/29552) - fix(auth): let internal users view search tools by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29542](https://github.com/BerriAI/litellm/pull/29542) - fix: missing mcp otel attributes by [@yassin-berriai](https://github.com/yassin-berriai) in [#29554](https://github.com/BerriAI/litellm/pull/29554) - fix(proxy): resolve managed video model ids for auth by [@shivamrawat1](https://github.com/shivamrawat1) in [#29545](https://github.com/BerriAI/litellm/pull/29545) - fix(key\_generate): allow team members to create keys on org-scoped teams by [@milan-berri](https://github.com/milan-berri) in [#29310](https://github.com/BerriAI/litellm/pull/29310) - test(pass-through): move Gemini pass-through tests to gemini-3.1-flash-lite by [@mateo-berri](https://github.com/mateo-berri) in [#29595](https://github.com/BerriAI/litellm/pull/29595) - Litellm oss staging 030626 by [@Sameerlite](https://github.com/Sameerlite) in [#29578](https://github.com/BerriAI/litellm/pull/29578) - Fix : a2a bugs 030626 by [@Sameerlite](https://github.com/Sameerlite) in [#29566](https://github.com/BerriAI/litellm/pull/29566) - \[internal copy of [#29533](https://github.com/BerriAI/litellm/issues/29533)] fix(anthropic/adapter): emit thinking block for reasoning\_content-only streaming chunks by [@mateo-berri](https://github.com/mateo-berri) in [#29600](https://github.com/BerriAI/litellm/pull/29600) - ci: reproduce default-Windows wheel install to guard MAX\_PATH by [@yuneng-berri](https://github.com/yuneng-berri) in [#29597](https://github.com/BerriAI/litellm/pull/29597) - fix(vertex): strip output\_config.effort for Vertex Claude models that reject it (Haiku 4.5) by [@mateo-berri](https://github.com/mateo-berri) in [#29585](https://github.com/BerriAI/litellm/pull/29585) - Litellm websocket improvements by [@Sameerlite](https://github.com/Sameerlite) in [#29563](https://github.com/BerriAI/litellm/pull/29563) - feat(arize/phoenix): OpenInference rendering parity — tool\_calls, cost, passthrough I/O, session/user, multimodal, cache tokens by [@milan-berri](https://github.com/milan-berri) in [#28800](https://github.com/BerriAI/litellm/pull/28800) - \[internal copy of [#29550](https://github.com/BerriAI/litellm/issues/29550)] fix: passthrough endpoints duplicate logs by [@mateo-berri](https://github.com/mateo-berri) in [#29598](https://github.com/BerriAI/litellm/pull/29598) - fix(ci): keep coverage rename green when a parallel node runs no tests by [@mateo-berri](https://github.com/mateo-berri) in [#29608](https://github.com/BerriAI/litellm/pull/29608) - test(vcr): close out the remaining VCR live-call leaks by [@mateo-berri](https://github.com/mateo-berri) in [#29603](https://github.com/BerriAI/litellm/pull/29603) - fix(key\_generate): exempt UI/CLI session tokens from the budget ceiling for team keys by [@yuneng-berri](https://github.com/yuneng-berri) in [#29612](https://github.com/BerriAI/litellm/pull/29612) - fix(realtime): allow null transcripts in stream logging payloads by [@milan-berri](https://github.com/milan-berri) in [#29625](https://github.com/BerriAI/litellm/pull/29625) - build(ui): migrate eslint to flat config + bump eslint-config-next to 16 by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29626](https://github.com/BerriAI/litellm/pull/29626) - fix(key\_generate): scope session-token team-key budget exemption to caller-supplied team\_id by [@yuneng-berri](https://github.com/yuneng-berri) in [#29641](https://github.com/BerriAI/litellm/pull/29641) - fix(proxy): disable proxy buffering on streaming SSE responses by [@mateo-berri](https://github.com/mateo-berri) in [#29557](https://github.com/BerriAI/litellm/pull/29557) - fix(mcp): gate /public/mcp\_hub strictly on litellm.public\_mcp\_servers by [@michelligabriele](https://github.com/michelligabriele) in [#27764](https://github.com/BerriAI/litellm/pull/27764) - ci(ui): frontend-lint job enforcing prettier + eslint on changed files by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29633](https://github.com/BerriAI/litellm/pull/29633) - fix(gemini): googleSearch + server-side tools and googleMaps JSON schema by [@Sameerlite](https://github.com/Sameerlite) in [#29582](https://github.com/BerriAI/litellm/pull/29582) - fix(proxy): passthrough 404 when SERVER\_ROOT\_PATH is set by [@Sameerlite](https://github.com/Sameerlite) in [#29658](https://github.com/BerriAI/litellm/pull/29658) - fix(gemini-realtime): use GA event names for Pipecat 1.3.x compatibility by [@Sameerlite](https://github.com/Sameerlite) in [#29662](https://github.com/BerriAI/litellm/pull/29662) - Litellm oss staging 040626 by [@Sameerlite](https://github.com/Sameerlite) in [#29671](https://github.com/BerriAI/litellm/pull/29671) - style(ui): prettier formatting pass over the dashboard by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29622](https://github.com/BerriAI/litellm/pull/29622) - chore: ignore prettier dashboard reformat in git blame by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29695](https://github.com/BerriAI/litellm/pull/29695) - fix(helm): Enable Backend Deployment to mount Gateway config.yaml by [@tin-berri](https://github.com/tin-berri) in [#29605](https://github.com/BerriAI/litellm/pull/29605) - \[internal copy of [#29277](https://github.com/BerriAI/litellm/issues/29277)] fix(proxy): add default=None to LiteLLM\_TeamMembership.litellm\_budget\_table by [@mateo-berri](https://github.com/mateo-berri) in [#29684](https://github.com/BerriAI/litellm/pull/29684) - test: make custom\_tokenizer proxy tests hermetic by [@yuneng-berri](https://github.com/yuneng-berri) in [#29643](https://github.com/BerriAI/litellm/pull/29643) - test(proxy): stop running real-DB tests in GitHub Actions unit jobs by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29700](https://github.com/BerriAI/litellm/pull/29700) - chore(ui): remove the bare-fetch lint rule by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29712](https://github.com/BerriAI/litellm/pull/29712) - Litellm jwt mapping virtualkeys by [@shivamrawat1](https://github.com/shivamrawat1) in [#28510](https://github.com/BerriAI/litellm/pull/28510) - refactor(ui): shared HTTP client + location-pinned fetch() lint rule by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29723](https://github.com/BerriAI/litellm/pull/29723) - fix(proxy): stop team BYOK model name corruption on model edit by [@yuneng-berri](https://github.com/yuneng-berri) in [#29731](https://github.com/BerriAI/litellm/pull/29731) - \[internal copy of [#29511](https://github.com/BerriAI/litellm/issues/29511)] feat(guardrails): add sensitive data routing to on-premise models by [@mateo-berri](https://github.com/mateo-berri) in [#29531](https://github.com/BerriAI/litellm/pull/29531) - fix(proxy/hooks): populate llm\_provider on internal rate-limit errors by [@mateo-berri](https://github.com/mateo-berri) in [#27707](https://github.com/BerriAI/litellm/pull/27707) - fix(vertex/anthropic): handle namespace tools and strip client\_metadata for codex compatibility by [@Sameerlite](https://github.com/Sameerlite) in [#29489](https://github.com/BerriAI/litellm/pull/29489) - Support OAuth M2M for Databricks Apps A2A agents by [@mateo-berri](https://github.com/mateo-berri) in [#29586](https://github.com/BerriAI/litellm/pull/29586) - fix: small CLAUDE.md nit by [@mateo-berri](https://github.com/mateo-berri) in [#29749](https://github.com/BerriAI/litellm/pull/29749) - fix(anthropic): route Claude Opus 4.8 through adaptive thinking by [@mateo-berri](https://github.com/mateo-berri) in [#29702](https://github.com/BerriAI/litellm/pull/29702) - fix(proxy): persist oauth2\_flow on MCP server registration by [@michelligabriele](https://github.com/michelligabriele) in [#29690](https://github.com/BerriAI/litellm/pull/29690) - \[internal copy of [#27491](https://github.com/BerriAI/litellm/issues/27491)] fix(realtime): Fix Realtime Audio Token Cost Tracking by [@mateo-berri](https://github.com/mateo-berri) in [#29722](https://github.com/BerriAI/litellm/pull/29722) - fix(galileo): use ingest traces API and standard logging payload by [@Sameerlite](https://github.com/Sameerlite) in [#29651](https://github.com/BerriAI/litellm/pull/29651) - fix(auth): expand all-team-models sentinel in can\_key\_call\_model for batch validation by [@Sameerlite](https://github.com/Sameerlite) in [#29746](https://github.com/BerriAI/litellm/pull/29746) - test(vcr): stop refreshing cassette TTL on read so cassettes lapse after 24h by [@mateo-berri](https://github.com/mateo-berri) in [#29784](https://github.com/BerriAI/litellm/pull/29784) - test(ci): record/replay OpenAI image gen so the spend E2E isn't outage-bound by [@mateo-berri](https://github.com/mateo-berri) in [#29787](https://github.com/BerriAI/litellm/pull/29787) - fix(ui): route MCP playground auth by oauth2 mode instead of token\_url by [@tin-berri](https://github.com/tin-berri) in [#29714](https://github.com/BerriAI/litellm/pull/29714) - refactor(ui): centralize proxy base URL resolution into tested resolver by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29793](https://github.com/BerriAI/litellm/pull/29793) - Litellm oss staging 050626 by [@Sameerlite](https://github.com/Sameerlite) in [#29774](https://github.com/BerriAI/litellm/pull/29774) - test(google): add google-genai SDK proxy integration tests by [@Sameerlite](https://github.com/Sameerlite) in [#29781](https://github.com/BerriAI/litellm/pull/29781) - fix(jwt): use resolved DB user\_id for spend on legacy email match by [@milan-berri](https://github.com/milan-berri) in [#29217](https://github.com/BerriAI/litellm/pull/29217) - feat(ui): generate dashboard API types from the proxy OpenAPI spec by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29816](https://github.com/BerriAI/litellm/pull/29816) - fix(proxy): drop deleted team BYOK model name from team.models by [@yuneng-berri](https://github.com/yuneng-berri) in [#29820](https://github.com/BerriAI/litellm/pull/29820) - feat(mcp): per-server env vars with global + per-user scopes by [@mateo-berri](https://github.com/mateo-berri) in [#28917](https://github.com/BerriAI/litellm/pull/28917) - refactor(ui): route behavior-preserving networking calls through apiClient by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29806](https://github.com/BerriAI/litellm/pull/29806) - fix(mcp): persist Tools-tab MCP OAuth token to DB by [@tin-berri](https://github.com/tin-berri) in [#29809](https://github.com/BerriAI/litellm/pull/29809) - fix(ui): require new expiration when regenerating an expired key by [@milan-berri](https://github.com/milan-berri) in [#29838](https://github.com/BerriAI/litellm/pull/29838) - refactor(ui): route query-building networking calls through apiClient by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29815](https://github.com/BerriAI/litellm/pull/29815) - Make the image-gen record/replay proxy report cache mode and per-request HIT/MISS by [@mateo-berri](https://github.com/mateo-berri) in [#29802](https://github.com/BerriAI/litellm/pull/29802) - feat(proxy): hot-reload .env in dev when running with --reload by [@mateo-berri](https://github.com/mateo-berri) in [#29783](https://github.com/BerriAI/litellm/pull/29783) - fix(ui): stop MCP playground tool calls from sending twice by [@tin-berri](https://github.com/tin-berri) in [#29821](https://github.com/BerriAI/litellm/pull/29821) - feat(fal\_ai): add Nano Banana / Gemini 2.5 Flash Image generation support by [@mateo-berri](https://github.com/mateo-berri) in [#29798](https://github.com/BerriAI/litellm/pull/29798) - Title: Fix managed batch cancel credential resolution by [@shivamrawat1](https://github.com/shivamrawat1) in [#29734](https://github.com/BerriAI/litellm/pull/29734) - Title: fix(proxy): resolve vector store file list credentials from team deployments by [@shivamrawat1](https://github.com/shivamrawat1) in [#29739](https://github.com/BerriAI/litellm/pull/29739) - refactor: convert AWS and GCP Terraform stacks into reusable modules … by [@yassin-berriai](https://github.com/yassin-berriai) in [#28103](https://github.com/BerriAI/litellm/pull/28103) - chore(ui): build ui for release by [@yuneng-berri](https://github.com/yuneng-berri) in [#29853](https://github.com/BerriAI/litellm/pull/29853) - fix(terraform/gcp): prompt for image\_registry in DeployStack one-click by [@yassin-berriai](https://github.com/yassin-berriai) in [#29852](https://github.com/BerriAI/litellm/pull/29852) - fix(terraform/gcp): abandon SQL user on destroy by [@yassin-berriai](https://github.com/yassin-berriai) in [#29855](https://github.com/BerriAI/litellm/pull/29855) - Extend the record/replay proxy to chat, embeddings, moderations, rerank, and Anthropic by [@mateo-berri](https://github.com/mateo-berri) in [#29847](https://github.com/BerriAI/litellm/pull/29847) - chore(deps): bump deps by [@yuneng-berri](https://github.com/yuneng-berri) in [#29860](https://github.com/BerriAI/litellm/pull/29860) - chore(ci): promote internal staging to main by [@yuneng-berri](https://github.com/yuneng-berri) in [#29861](https://github.com/BerriAI/litellm/pull/29861) - fix: 400 on Anthropic context overflow; seed identity on failed auth by [@yassin-berriai](https://github.com/yassin-berriai) in [#29848](https://github.com/BerriAI/litellm/pull/29848) - chore(ci): promote internal staging to main by [@yuneng-berri](https://github.com/yuneng-berri) in [#29862](https://github.com/BerriAI/litellm/pull/29862) - chore(release): patch v1.89.0-rc.1 with [#30064](https://github.com/BerriAI/litellm/issues/30064) (Claude Fable 5) for v1.89.0-rc.2 by [@mateo-berri](https://github.com/mateo-berri) in [#30143](https://github.com/BerriAI/litellm/pull/30143) **Full Changelog**: <https://github.com/BerriAI/litellm/compare/v1.88.0...v1.89.0> ### [`v1.88.2`](https://github.com/BerriAI/litellm/releases/tag/v1.88.2) [Compare Source](https://github.com/BerriAI/litellm/compare/v1.88.1...v1.88.2) #### Verify Docker Image Signature All LiteLLM Docker images are signed with [cosign](https://docs.sigstore.dev/cosign/overview/). Every release is signed with the same key introduced in [commit `0112e53`](https://github.com/BerriAI/litellm/commit/0112e53046018d726492c814b3644b7d376029d0). **Verify using the pinned commit hash (recommended):** A commit hash is cryptographically immutable, so this is the strongest way to ensure you are using the original signing key: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/0112e53046018d726492c814b3644b7d376029d0/cosign.pub \ ghcr.io/berriai/litellm:v1.88.2 ``` **Verify using the release tag (convenience):** Tags are protected in this repository and resolve to the same key. This option is easier to read but relies on tag protection rules: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/v1.88.2/cosign.pub \ ghcr.io/berriai/litellm:v1.88.2 ``` Expected output: ``` The following checks were performed on each of these signatures: - The cosign claims were validated - The signatures were verified against the specified public key ``` *** #### What's Changed - chore(release): backport Fable 5, batch-file auth, CrowdStrike AIDR, Mantle Responses SigV4, and NetApp streaming-cost fix to stable/1.88.x and cut 1.88.2 by [@mateo-berri](https://github.com/mateo-berri) in [#30144](https://github.com/BerriAI/litellm/pull/30144) - chore(release): backport DB-resilience, passthrough, model-info, budget, and deps fixes to stable/1.88.x by [@yuneng-berri](https://github.com/yuneng-berri) in [#30408](https://github.com/BerriAI/litellm/pull/30408) **Full Changelog**: <https://github.com/BerriAI/litellm/compare/v1.88.1...v1.88.2> ### [`v1.88.1`](https://github.com/BerriAI/litellm/releases/tag/v1.88.1) [Compare Source](https://github.com/BerriAI/litellm/compare/v1.88.0...v1.88.1) #### Verify Docker Image Signature All LiteLLM Docker images are signed with [cosign](https://docs.sigstore.dev/cosign/overview/). Every release is signed with the same key introduced in [commit `0112e53`](https://github.com/BerriAI/litellm/commit/0112e53046018d726492c814b3644b7d376029d0). **Verify using the pinned commit hash (recommended):** A commit hash is cryptographically immutable, so this is the strongest way to ensure you are using the original signing key: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/0112e53046018d726492c814b3644b7d376029d0/cosign.pub \ ghcr.io/berriai/litellm:v1.88.1 ``` **Verify using the release tag (convenience):** Tags are protected in this repository and resolve to the same key. This option is easier to read but relies on tag protection rules: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/v1.88.1/cosign.pub \ ghcr.io/berriai/litellm:v1.88.1 ``` Expected output: ``` The following checks were performed on each of these signatures: - The cosign claims were validated - The signatures were verified against the specified public key ``` *** #### What's Changed - build(deps): bump pyjwt to 2.13.0 and ws override to 8.20.1 (1.88.x) by [@yuneng-berri](https://github.com/yuneng-berri) in [#29987](https://github.com/BerriAI/litellm/pull/29987) - chore(release): bump version to 1.88.1 by [@yuneng-berri](https://github.com/yuneng-berri) in [#29989](https://github.com/BerriAI/litellm/pull/29989) **Full Changelog**: <https://github.com/BerriAI/litellm/compare/v1.88.0...v1.88.1> ### [`v1.88.0`](https://github.com/BerriAI/litellm/releases/tag/v1.88.0) [Compare Source](https://github.com/BerriAI/litellm/compare/v1.87.3...v1.88.0) #### Verify Docker Image Signature All LiteLLM Docker images are signed with [cosign](https://docs.sigstore.dev/cosign/overview/). Every release is signed with the same key introduced in [commit `0112e53`](https://github.com/BerriAI/litellm/commit/0112e53046018d726492c814b3644b7d376029d0). **Verify using the pinned commit hash (recommended):** A commit hash is cryptographically immutable, so this is the strongest way to ensure you are using the original signing key: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/0112e53046018d726492c814b3644b7d376029d0/cosign.pub \ ghcr.io/berriai/litellm:v1.88.0 ``` **Verify using the release tag (convenience):** Tags are protected in this repository and resolve to the same key. This option is easier to read but relies on tag protection rules: ```bash cosign verify \ --key https://raw.githubusercontent.com/BerriAI/litellm/v1.88.0/cosign.pub \ ghcr.io/berriai/litellm:v1.88.0 ``` Expected output: ``` The following checks were performed on each of these signatures: - The cosign claims were validated - The signatures were verified against the specified public key ``` *** #### What's Changed - fix(proxy): gate team allowed\_passthrough\_routes to proxy admins by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28097](https://github.com/BerriAI/litellm/pull/28097) - fix(tests): stabilize image-edit VCR cassettes to stop live gpt-image-1 spend by [@mateo-berri](https://github.com/mateo-berri) in [#28110](https://github.com/BerriAI/litellm/pull/28110) - fix(bedrock/cohere): send embedding\_types as JSON array, not string by [@ishaan-berri](https://github.com/ishaan-berri) in [#28172](https://github.com/BerriAI/litellm/pull/28172) - fix(tests): migrate realtime + rerank tests off shut-down upstream models by [@yuneng-berri](https://github.com/yuneng-berri) in [#28191](https://github.com/BerriAI/litellm/pull/28191) - fix(caching): replay openai/responses bridge cache hits as chat streams by [@Sameerlite](https://github.com/Sameerlite) in [#28158](https://github.com/BerriAI/litellm/pull/28158) - Litellm oss staging by [@Sameerlite](https://github.com/Sameerlite) in [#28161](https://github.com/BerriAI/litellm/pull/28161) - feat(prometheus): add user\_email and user\_alias to user budget metrics by [@Sameerlite](https://github.com/Sameerlite) in [#28155](https://github.com/BerriAI/litellm/pull/28155) - test(callbacks): harden flaky proxy callback-leak detector by [@yuneng-berri](https://github.com/yuneng-berri) in [#28195](https://github.com/BerriAI/litellm/pull/28195) - fix(bedrock): sanitize batch metadata to prevent Pydantic ValidationError by [@mateo-berri](https://github.com/mateo-berri) in [#28202](https://github.com/BerriAI/litellm/pull/28202) - fix(deepseek): use native /anthropic/v1/messages endpoint and sanitize tools by [@mateo-berri](https://github.com/mateo-berri) in [#28200](https://github.com/BerriAI/litellm/pull/28200) - feat(ui): add Interactions API endpoint to playground with SSE streaming by [@Sameerlite](https://github.com/Sameerlite) in [#28156](https://github.com/BerriAI/litellm/pull/28156) - fix(proxy): decode bytes and pass-through SSE for Google-native streamGenerateContent ([#27444](https://github.com/BerriAI/litellm/issues/27444)) by [@Sameerlite](https://github.com/Sameerlite) in [#28213](https://github.com/BerriAI/litellm/pull/28213) - refactor(bedrock/sagemaker): switch to lazy loading for response stre… by [@harish-berri](https://github.com/harish-berri) in [#28189](https://github.com/BerriAI/litellm/pull/28189) - \[Refactor] UI - Spend Logs: consolidate filter state and extract components by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#25847](https://github.com/BerriAI/litellm/pull/25847) - fix(tests): replace shut-down gpt-4o-audio-preview with gpt-audio-1.5 by [@yuneng-berri](https://github.com/yuneng-berri) in [#28281](https://github.com/BerriAI/litellm/pull/28281) - chore(ci): bump versions by [@yuneng-berri](https://github.com/yuneng-berri) in [#28287](https://github.com/BerriAI/litellm/pull/28287) - feat: propagate team\_id and team\_alias to all child OTEL spans by [@yassin-berriai](https://github.com/yassin-berriai) in [#28273](https://github.com/BerriAI/litellm/pull/28273) - Day 0 support : Gemini 3.5 Flash by [@Sameerlite](https://github.com/Sameerlite) in [#28268](https://github.com/BerriAI/litellm/pull/28268) - Gemini managed agents support by [@Sameerlite](https://github.com/Sameerlite) in [#28270](https://github.com/BerriAI/litellm/pull/28270) - chore(ci): promote internal staging to main by [@yuneng-berri](https://github.com/yuneng-berri) in [#28292](https://github.com/BerriAI/litellm/pull/28292) - feat(gemini): add gemini-3.1-flash-lite model cost map by [@Sameerlite](https://github.com/Sameerlite) in [#28320](https://github.com/BerriAI/litellm/pull/28320) - fix(spend\_counter): seed Redis counter via SET NX to prevent cross-pod double-seed by [@milan-berri](https://github.com/milan-berri) in [#27854](https://github.com/BerriAI/litellm/pull/27854) - fix(proxy): normalize batch file IDs before ManagedObjectTable write by [@Sameerlite](https://github.com/Sameerlite) in [#28339](https://github.com/BerriAI/litellm/pull/28339) - fix(router): use forwarded model\_id for native Azure container IDs by [@Sameerlite](https://github.com/Sameerlite) in [#27921](https://github.com/BerriAI/litellm/pull/27921) - fix(ui): restore log filter loading indicator by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28282](https://github.com/BerriAI/litellm/pull/28282) - test(e2e): migrate runner to uv, add All Proxy Models key test by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28313](https://github.com/BerriAI/litellm/pull/28313) - feat(ui): team passthrough routes create parity + edit load fix by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28098](https://github.com/BerriAI/litellm/pull/28098) - fix(mcp): JWT on tools/list and REST tools/call server resolution by [@Sameerlite](https://github.com/Sameerlite) in [#28227](https://github.com/BerriAI/litellm/pull/28227) - feat(interactions): migrate to Google Interactions API steps schema (May 2026) by [@Sameerlite](https://github.com/Sameerlite) in [#28153](https://github.com/BerriAI/litellm/pull/28153) - test(ui-e2e): admin key creation with a specific proxy model by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28365](https://github.com/BerriAI/litellm/pull/28365) - fix(vertex\_ai): omit function\_call id on Vertex Gemini 3.5+ tool turns by [@Sameerlite](https://github.com/Sameerlite) in [#28324](https://github.com/BerriAI/litellm/pull/28324) - feat(mcp): allow native MCP OAuth support for cursor by [@Sameerlite](https://github.com/Sameerlite) in [#28327](https://github.com/BerriAI/litellm/pull/28327) - fix(interactions): never drop streamed text deltas; always emit terminal completion by [@mateo-berri](https://github.com/mateo-berri) in [#28394](https://github.com/BerriAI/litellm/pull/28394) - fix(proxy): expose Prisma idle/connect timeout + extra DB URL params by [@yassin-berriai](https://github.com/yassin-berriai) in [#28395](https://github.com/BerriAI/litellm/pull/28395) - Litellm oss staging 1 by [@Sameerlite](https://github.com/Sameerlite) in [#28337](https://github.com/BerriAI/litellm/pull/28337) - fix: serialize guardrail\_response to JSON in OTEL traces by [@yassin-berriai](https://github.com/yassin-berriai) in [#28362](https://github.com/BerriAI/litellm/pull/28362) - chore(ci): merge dev branch by [@yuneng-berri](https://github.com/yuneng-berri) in [#28314](https://github.com/BerriAI/litellm/pull/28314) - test(realtime): expect session.created as xAI realtime initial event by [@yuneng-berri](https://github.com/yuneng-berri) in [#28424](https://github.com/BerriAI/litellm/pull/28424) - feat(tests): behavior-pinning harness + Key Tier-1 matrix by [@yuneng-berri](https://github.com/yuneng-berri) in [#28321](https://github.com/BerriAI/litellm/pull/28321) - fix(proxy): hydrate wildcard discovery credentials ([#28284](https://github.com/BerriAI/litellm/issues/28284)) - CCI Run by [@yuneng-berri](https://github.com/yuneng-berri) in [#28419](https://github.com/BerriAI/litellm/pull/28419) - Litellm oss staging 04 21 2026 2 by [@Sameerlite](https://github.com/Sameerlite) in [#26569](https://github.com/BerriAI/litellm/pull/26569) - chore(ci): merge dev branch by [@yuneng-berri](https://github.com/yuneng-berri) in [#28290](https://github.com/BerriAI/litellm/pull/28290) - fix(vertex\_gemma): strip `context_management` from request body by [@mateo-berri](https://github.com/mateo-berri) in [#28438](https://github.com/BerriAI/litellm/pull/28438) - fix(logging): recalculate cost after router retry failures by [@milan-berri](https://github.com/milan-berri) in [#28476](https://github.com/BerriAI/litellm/pull/28476) - fix(otel): emit guardrail span on violation, surface status + categories by [@yassin-berriai](https://github.com/yassin-berriai) in [#28364](https://github.com/BerriAI/litellm/pull/28364) - test(proxy): behavior-pinning matrix for team management endpoints by [@yuneng-berri](https://github.com/yuneng-berri) in [#28441](https://github.com/BerriAI/litellm/pull/28441) - test(vertex\_ai): tolerate transient 500 in google maps grounding test by [@yuneng-berri](https://github.com/yuneng-berri) in [#28503](https://github.com/BerriAI/litellm/pull/28503) - fix(docker): restore npm to non\_root builder image by [@yuneng-berri](https://github.com/yuneng-berri) in [#28519](https://github.com/BerriAI/litellm/pull/28519) - chore(ci): bump deps by [@yuneng-berri](https://github.com/yuneng-berri) in [#28524](https://github.com/BerriAI/litellm/pull/28524) - build(deps-dev): bump black to 26.3.1 and apply formatting by [@yuneng-berri](https://github.com/yuneng-berri) in [#28525](https://github.com/BerriAI/litellm/pull/28525) - chore(deps): bump deps by [@yuneng-berri](https://github.com/yuneng-berri) in [#28528](https://github.com/BerriAI/litellm/pull/28528) - test(e2e): forward LITELLM\_LICENSE to UI e2e proxy by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28398](https://github.com/BerriAI/litellm/pull/28398) - Add granian as a ASGI compliant web server. Provider better throughput stability, by [@harish-berri](https://github.com/harish-berri) in [#26027](https://github.com/BerriAI/litellm/pull/26027) - Fix conflicts and UI by [@Sameerlite](https://github.com/Sameerlite) in [#28477](https://github.com/BerriAI/litellm/pull/28477) - Add error\_description and hint for oauth flows by [@Sameerlite](https://github.com/Sameerlite) in [#28471](https://github.com/BerriAI/litellm/pull/28471) - feat(mcp): Add tool call and tool list support via UI for Oauth mcps by [@Sameerlite](https://github.com/Sameerlite) in [#28454](https://github.com/BerriAI/litellm/pull/28454) - feat(proxy): persist allowlisted OIDC claims in CLI SSO poll by [@Sameerlite](https://github.com/Sameerlite) in [#28463](https://github.com/BerriAI/litellm/pull/28463) - fix(responses): use OpenAI SSEDecoder for Responses API streaming by [@Sameerlite](https://github.com/Sameerlite) in [#28566](https://github.com/BerriAI/litellm/pull/28566) - Litellm oss staging 2 by [@Sameerlite](https://github.com/Sameerlite) in [#28582](https://github.com/BerriAI/litellm/pull/28582) - \[internal copy of [#28269](https://github.com/BerriAI/litellm/issues/28269)] Codex cli jwt team alias by [@mateo-berri](https://github.com/mateo-berri) in [#28621](https://github.com/BerriAI/litellm/pull/28621) - fix(check\_licenses): read PEP 639 license-expression metadata by [@yuneng-berri](https://github.com/yuneng-berri) in [#28529](https://github.com/BerriAI/litellm/pull/28529) - test(proxy): behavior-pinning matrix for tier-2/3 key + team management endpoints by [@yuneng-berri](https://github.com/yuneng-berri) in [#28620](https://github.com/BerriAI/litellm/pull/28620) - chore(test): remove dead old Playwright e2e suite by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28632](https://github.com/BerriAI/litellm/pull/28632) - fix(sagemaker): send native Cohere embed payload to Cohere SageMaker endpoints by [@milan-berri](https://github.com/milan-berri) in [#28613](https://github.com/BerriAI/litellm/pull/28613) - style: apply black formatting to fix lint CI (LIT-3274) ([#28639](https://github.com/BerriAI/litellm/issues/28639)) by [@krrish-berri-2](https://github.com/krrish-berri-2) in [#28641](https://github.com/BerriAI/litellm/pull/28641) - fix(bedrock): decouple STS region from Bedrock aws\_region\_name by [@milan-berri](https://github.com/milan-berri) in [#28245](https://github.com/BerriAI/litellm/pull/28245) - test(streaming): tolerate Vertex 429 wrapped in MidStreamFallbackError by [@yuneng-berri](https://github.com/yuneng-berri) in [#28669](https://github.com/BerriAI/litellm/pull/28669) - feat(guardrails): add Microsoft Purview DLP guardrail by [@Sameerlite](https://github.com/Sameerlite) in [#24966](https://github.com/BerriAI/litellm/pull/24966) - fix(mcp): forward upstream initialize instructions on cold gateway init by [@milan-berri](https://github.com/milan-berri) in [#28231](https://github.com/BerriAI/litellm/pull/28231) - chore(ci): promote internal staging to main by [@yuneng-berri](https://github.com/yuneng-berri) in [#28680](https://github.com/BerriAI/litellm/pull/28680) - CI: copy of [#25177](https://github.com/BerriAI/litellm/issues/25177) (OCI GenAI: embeddings, streaming/reasoning fixes, model catalog) by [@mateo-berri](https://github.com/mateo-berri) in [#28223](https://github.com/BerriAI/litellm/pull/28223) - Encrypt callback\_vars in key/team metadata in DB by [@Michael-RZ-Berri](https://github.com/Michael-RZ-Berri) in [#27141](https://github.com/BerriAI/litellm/pull/27141) - perf: reduce per-request and per-chunk overhead across Anthropic streaming hot paths by [@yassin-berriai](https://github.com/yassin-berriai) in [#28289](https://github.com/BerriAI/litellm/pull/28289) - feat(azure): add Speech STT config support by [@ishaan-berri](https://github.com/ishaan-berri) in [#27482](https://github.com/BerriAI/litellm/pull/27482) - test(proxy): phase-4 payload behavior pinning for tier-2/3 key + team management endpoints by [@yuneng-berri](https://github.com/yuneng-berri) in [#28681](https://github.com/BerriAI/litellm/pull/28681) - feat(prometheus): emit per-token-type detail metrics (LIT-3220) ([#28372](https://github.com/BerriAI/litellm/issues/28372)) by [@ishaan-berri](https://github.com/ishaan-berri) in [#28378](https://github.com/BerriAI/litellm/pull/28378) - fix(otel): stamp http.response.status\_code on all error responses by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28405](https://github.com/BerriAI/litellm/pull/28405) - chore(ui): build ui by [@yuneng-berri](https://github.com/yuneng-berri) in [#28707](https://github.com/BerriAI/litellm/pull/28707) - fix(helm): drop main- prefix from default image tag by [@yuneng-berri](https://github.com/yuneng-berri) in [#28710](https://github.com/BerriAI/litellm/pull/28710) - test(model\_prices): allow audio\_transcription\_config in schema by [@yuneng-berri](https://github.com/yuneng-berri) in [#28708](https://github.com/BerriAI/litellm/pull/28708) - chore(ci): promote internal staging to main by [@yuneng-berri](https://github.com/yuneng-berri) in [#28709](https://github.com/BerriAI/litellm/pull/28709) - fix(team): refresh team cache on team\_model\_add/delete (LIT-3244) by [@yuneng-berri](https://github.com/yuneng-berri) in [#28683](https://github.com/BerriAI/litellm/pull/28683) - fix(ui/add-model): stop vertex\_ai-anthropic\_models from leaking into Anthropic dropdown by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28723](https://github.com/BerriAI/litellm/pull/28723) - Fix spend logs v2 route permissions by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28705](https://github.com/BerriAI/litellm/pull/28705) - fix(proxy): Bedrock Knowledge Base pass-through: preserve SigV4 headers and signed request body by [@milan-berri](https://github.com/milan-berri) in [#27526](https://github.com/BerriAI/litellm/pull/27526) - chore(tests): migrate Bedrock CI to AWS account [`9412775`](https://github.com/BerriAI/litellm/commit/941277531214) by [@mateo-berri](https://github.com/mateo-berri) in [#28728](https://github.com/BerriAI/litellm/pull/28728) - fix(otel): export SERVER span on management-endpoint success without http\_request by [@yassin-berriai](https://github.com/yassin-berriai) in [#28794](https://github.com/BerriAI/litellm/pull/28794) - chore(ci): merge dev branch by [@yuneng-berri](https://github.com/yuneng-berri) in [#28801](https://github.com/BerriAI/litellm/pull/28801) - chore(ci): merge dev branch by [@yuneng-berri](https://github.com/yuneng-berri) in [#28657](https://github.com/BerriAI/litellm/pull/28657) - fix(ui): show 2-decimal precision for max\_budget on key overview by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28809](https://github.com/BerriAI/litellm/pull/28809) - feat(proxy): allow `llm_api_routes` virtual keys to list MCP servers by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28442](https://github.com/BerriAI/litellm/pull/28442) - chore(ci): merge dev branch by [@yuneng-berri](https://github.com/yuneng-berri) in [#28807](https://github.com/BerriAI/litellm/pull/28807) - fix(team): keep team\_alias cache in sync on \_cache\_team\_object writes by [@yuneng-berri](https://github.com/yuneng-berri) in [#28737](https://github.com/BerriAI/litellm/pull/28737) - chore(ci): merge dev branch by [@yuneng-berri](https://github.com/yuneng-berri) in [#28822](https://github.com/BerriAI/litellm/pull/28822) - ci: daily oss-agent-shin canonical branch by [@ishaan-berri](https://github.com/ishaan-berri) in [#28829](https://github.com/BerriAI/litellm/pull/28829) - test(proxy): add harness for proxy\_server.py behavior-pinning by [@yuneng-berri](https://github.com/yuneng-berri) in [#28827](https://github.com/BerriAI/litellm/pull/28827) - feat(openai): apply regional-processing cost uplift for EU/US data residency by [@mateo-berri](https://github.com/mateo-berri) in [#28626](https://github.com/BerriAI/litellm/pull/28626) - chore(admin-ui): regenerate static export with trailingSlash: true by [@mateo-berri](https://github.com/mateo-berri) in [#28112](https://github.com/BerriAI/litellm/pull/28112) - fix(azure): preserve AD token refresh in v1 OpenAI client path by [@mateo-berri](https://github.com/mateo-berri) in [#28627](https://github.com/BerriAI/litellm/pull/28627) - fix(ui): route API Reference back to query-param page by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28726](https://github.com/BerriAI/litellm/pull/28726) - fix(model-edit): allow clearing custom pricing on wildcard models by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28719](https://github.com/BerriAI/litellm/pull/28719) - fix(tests/vcr): make Redis cassette cache replay deterministically (zero VCR misses on consecutive runs) by [@mateo-berri](https://github.com/mateo-berri) in [#28826](https://github.com/BerriAI/litellm/pull/28826) - fix(proxy): strip LiteLLM policy tracking from OpenAI batch metadata by [@shivamrawat1](https://github.com/shivamrawat1) in [#28425](https://github.com/BerriAI/litellm/pull/28425) - Litellm OpenAI double prefix bug by [@shivamrawat1](https://github.com/shivamrawat1) in [#28661](https://github.com/BerriAI/litellm/pull/28661) - Litellm oss staging 250526 by [@Sameerlite](https://github.com/Sameerlite) in [#28770](https://github.com/BerriAI/litellm/pull/28770) - fix(bedrock): align toolUse/toolSpec names and allow hyphens by [@Sameerlite](https://github.com/Sameerlite) in [#28874](https://github.com/BerriAI/litellm/pull/28874) - fix(realtime): send TEXT frames and valid guardrail session.update by [@Sameerlite](https://github.com/Sameerlite) in [#28848](https://github.com/BerriAI/litellm/pull/28848) - fix(mcp): extend key access-group union to MCP servers by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28890](https://github.com/BerriAI/litellm/pull/28890) - fix(galileo): support hosted v2 spans API and string output extraction by [@Sameerlite](https://github.com/Sameerlite) in [#28771](https://github.com/BerriAI/litellm/pull/28771) - fix(proxy): exclude proxy\_server\_request from its own body snapshot by [@michelligabriele](https://github.com/michelligabriele) in [#28618](https://github.com/BerriAI/litellm/pull/28618) - \[Feat] Add tool calling support for gemini and vertex ai live api by [@Sameerlite](https://github.com/Sameerlite) in [#26590](https://github.com/BerriAI/litellm/pull/26590) - refactor(ui): remove dead App Router scaffolding in (dashboard)/\* by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28891](https://github.com/BerriAI/litellm/pull/28891) - fix(docker): use system Node in componentized builders + retry apk add by [@yassin-berriai](https://github.com/yassin-berriai) in [#28888](https://github.com/BerriAI/litellm/pull/28888) - docs(agents): require consent before writing new third-party names by [@yuneng-berri](https://github.com/yuneng-berri) in [#28908](https://github.com/BerriAI/litellm/pull/28908) - refactor(ui): extract auth state into AuthContext by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28910](https://github.com/BerriAI/litellm/pull/28910) - fix(mcp): resolve team.access\_group\_ids → MCP servers by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28997](https://github.com/BerriAI/litellm/pull/28997) - test(ui): e2e cover team model edit + admin identity in navbar by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#28652](https://github.com/BerriAI/litellm/pull/28652) - test(e2e): cover add-fallback flow in Router Settings by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29069](https://github.com/BerriAI/litellm/pull/29069) - test(e2e): cover Team-BYOK add-model flow as proxy admin by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29068](https://github.com/BerriAI/litellm/pull/29068) - fix(containers): record ownership for service-account keys + fix Prisma Json serialization by [@Sameerlite](https://github.com/Sameerlite) in [#28990](https://github.com/BerriAI/litellm/pull/28990) - test(e2e): cover add-MCP-server flow via discovery → custom form by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29070](https://github.com/BerriAI/litellm/pull/29070) - test(e2e): cover AI Hub make-public flow and public model\_hub\_table by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29071](https://github.com/BerriAI/litellm/pull/29071) - \[internal copy of [#28877](https://github.com/BerriAI/litellm/issues/28877)] feat: add support for claude code goal mode for bedrock opus output config by [@mateo-berri](https://github.com/mateo-berri) in [#28898](https://github.com/BerriAI/litellm/pull/28898) - feat(guardrails): wire apply\_guardrail into proxy logging callbacks by [@Sameerlite](https://github.com/Sameerlite) in [#28970](https://github.com/BerriAI/litellm/pull/28970) - chore(ci): merge dev brach by [@yuneng-berri](https://github.com/yuneng-berri) in [#29192](https://github.com/BerriAI/litellm/pull/29192) - perf(streaming): cut per-chunk overhead \~30% on Anthropic + Bedrock hot path by [@yassin-berriai](https://github.com/yassin-berriai) in [#28720](https://github.com/BerriAI/litellm/pull/28720) - fix(proxy): enforce tag budgets for key-level tags by [@Sameerlite](https://github.com/Sameerlite) in [#29108](https://github.com/BerriAI/litellm/pull/29108) - fix(vertex-ai): use DB credentials in video handlers + implement Veo video edit by [@Sameerlite](https://github.com/Sameerlite) in [#29098](https://github.com/BerriAI/litellm/pull/29098) - fix(datadog): drain cost-management queue + opt-in FinOps tag allowlist by [@michelligabriele](https://github.com/michelligabriele) in [#28487](https://github.com/BerriAI/litellm/pull/28487) - feat(helm): split per-component ServiceAccounts for gateway, backend, and UI by [@yassin-berriai](https://github.com/yassin-berriai) in [#28712](https://github.com/BerriAI/litellm/pull/28712) - chore(ci): bump deps ([#29208](https://github.com/BerriAI/litellm/issues/29208)) by [@yuneng-berri](https://github.com/yuneng-berri) in [#29226](https://github.com/BerriAI/litellm/pull/29226) - fix(tests/vcr): mint Google OAuth tokens live to prevent stale-token replay by [@yuneng-berri](https://github.com/yuneng-berri) in [#29229](https://github.com/BerriAI/litellm/pull/29229) - chore(cookbook): bump Go directive to 1.26.3 in gollem example by [@yuneng-berri](https://github.com/yuneng-berri) in [#29234](https://github.com/BerriAI/litellm/pull/29234) - chore(ci): bump version by [@yuneng-berri](https://github.com/yuneng-berri) in [#29242](https://github.com/BerriAI/litellm/pull/29242) - feat(anthropic): add Claude Opus 4.8 and prune reasoning-effort flags by [@mateo-berri](https://github.com/mateo-berri) in [#29238](https://github.com/BerriAI/litellm/pull/29238) - chore(ci): promote internal staging to main by [@yuneng-berri](https://github.com/yuneng-berri) in [#29243](https://github.com/BerriAI/litellm/pull/29243) - fix(ci): restore real Bedrock batch S3 bucket/role in oai\_misc\_config by [@mateo-berri](https://github.com/mateo-berri) in [#29245](https://github.com/BerriAI/litellm/pull/29245) - fix(guardrails): persist disable\_global\_guardrails on keys by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29233](https://github.com/BerriAI/litellm/pull/29233) - test(e2e): cover Team Admin view + member + key flows by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29072](https://github.com/BerriAI/litellm/pull/29072) - docs: hand-written CLAUDE.md; remove AGENTS.md, point GEMINI.md at it by [@mateo-berri](https://github.com/mateo-berri) in [#29252](https://github.com/BerriAI/litellm/pull/29252) - fix(teams): expose keys\_count on /v2/team/list and wire UI Resources badge by [@michelligabriele](https://github.com/michelligabriele) in [#28502](https://github.com/BerriAI/litellm/pull/28502) - fix(anthropic): stop injecting unsupported output\_config.effort=xhigh for Claude Code on Sonnet/Opus 4.6 by [@mateo-berri](https://github.com/mateo-berri) in [#29304](https://github.com/BerriAI/litellm/pull/29304) - test(e2e): cover Internal Viewer nav, key, and team-info gating by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29075](https://github.com/BerriAI/litellm/pull/29075) - test(e2e): cover Internal User key modal, team info, key page by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29074](https://github.com/BerriAI/litellm/pull/29074) - test(e2e): cover navbar Logout flow as proxy admin by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29076](https://github.com/BerriAI/litellm/pull/29076) - fix(mcp): resolve key.access\_group\_ids → MCP servers (ungated) by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29195](https://github.com/BerriAI/litellm/pull/29195) - fix(router): enforce deployment budgets for dynamically added models by [@Sameerlite](https://github.com/Sameerlite) in [#29273](https://github.com/BerriAI/litellm/pull/29273) - fix(proxy): map stripped batch body.model to proxy alias for auth by [@Sameerlite](https://github.com/Sameerlite) in [#29264](https://github.com/BerriAI/litellm/pull/29264) - feat(mcp): support stateless and stateful clients via session-id routing by [@Sameerlite](https://github.com/Sameerlite) in [#26857](https://github.com/BerriAI/litellm/pull/26857) - fix(bedrock): support tool search results + chat annotations by [@Sameerlite](https://github.com/Sameerlite) in [#29120](https://github.com/BerriAI/litellm/pull/29120) - fix(mcp): ignore stale ids on key save by [@Sameerlite](https://github.com/Sameerlite) in [#29128](https://github.com/BerriAI/litellm/pull/29128) - feat(a2a): well-known agent-card discovery + LangGraph Platform mode by [@Sameerlite](https://github.com/Sameerlite) in [#28860](https://github.com/BerriAI/litellm/pull/28860) - fix(proxy): link passthrough success spans to the SERVER root OTEL span by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29315](https://github.com/BerriAI/litellm/pull/29315) - \[internal copy of [#29089](https://github.com/BerriAI/litellm/issues/29089)] fix: duplicate claude code traces by [@mateo-berri](https://github.com/mateo-berri) in [#29311](https://github.com/BerriAI/litellm/pull/29311) - feat(otel): typed semconv-aligned OpenTelemetry instrumentation by [@yassin-berriai](https://github.com/yassin-berriai) in [#28909](https://github.com/BerriAI/litellm/pull/28909) - tests(proxy\_server): surface current behavior in tests by [@yuneng-berri](https://github.com/yuneng-berri) in [#29309](https://github.com/BerriAI/litellm/pull/29309) - test(e2e): cover Internal User create-key flow when in no teams by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29083](https://github.com/BerriAI/litellm/pull/29083) - test(e2e): assert internal-user navbar identity is scoped to that user by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29077](https://github.com/BerriAI/litellm/pull/29077) - feat(otel): add team\_metadata, http.route, and model names to inference spans by [@yassin-berriai](https://github.com/yassin-berriai) in [#29319](https://github.com/BerriAI/litellm/pull/29319) - feat(context\_management): compact\_20260112 polyfill for non-Anthropic providers by [@Sameerlite](https://github.com/Sameerlite) in [#28868](https://github.com/BerriAI/litellm/pull/28868) - feat(enterprise): add RESEND\_FROM\_EMAIL for self-hosted Resend sends by [@shivamrawat1](https://github.com/shivamrawat1) in [#28830](https://github.com/BerriAI/litellm/pull/28830) - Revert Bedrock CI back to the reactivated AWS account ([`8886022`](https://github.com/BerriAI/litellm/commit/888602223428)) by [@mateo-berri](https://github.com/mateo-berri) in [#29326](https://github.com/BerriAI/litellm/pull/29326) - fix(mcp): preserve source\_url in GET /v1/mcp/server list responses by [@shivamrawat1](https://github.com/shivamrawat1) in [#29249](https://github.com/BerriAI/litellm/pull/29249) - fix(mcp): preserve omitted fields on PUT /v1/mcp/server partial updates by [@shivamrawat1](https://github.com/shivamrawat1) in [#29253](https://github.com/BerriAI/litellm/pull/29253) - fix(ci): make litellm\_internal\_staging green (logging test + Bedrock Opus 4.7 self-heal) by [@mateo-berri](https://github.com/mateo-berri) in [#29344](https://github.com/BerriAI/litellm/pull/29344) - refactor(proxy/auth): normalize Bearer prefix in safe-hash helper by [@yuneng-berri](https://github.com/yuneng-berri) in [#29343](https://github.com/BerriAI/litellm/pull/29343) - test(reasoning-effort-grid): cover Claude Opus 4.8 across provider routes by [@mateo-berri](https://github.com/mateo-berri) in [#29327](https://github.com/BerriAI/litellm/pull/29327) - fix(guardrails): return HTTP 400 for litellm content filter blocks by [@shivamrawat1](https://github.com/shivamrawat1) in [#28418](https://github.com/BerriAI/litellm/pull/28418) - fix(proxy): restrict vector store index create/delete to proxy admins by [@shivamrawat1](https://github.com/shivamrawat1) in [#29202](https://github.com/BerriAI/litellm/pull/29202) - feat(pass\_through): extend passthrough\_managed\_object\_ids to Azure by [@Sameerlite](https://github.com/Sameerlite) in [#29160](https://github.com/BerriAI/litellm/pull/29160) - fix(proxy): enforce allowed\_passthrough\_routes for auth=true pass-thr… by [@shivamrawat1](https://github.com/shivamrawat1) in [#29256](https://github.com/BerriAI/litellm/pull/29256) - feat(mcp/auth): additive key access-group grants + opt-in member assignment by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29313](https://github.com/BerriAI/litellm/pull/29313) - fix(reset\_budget): write only {spend, budget\_reset\_at} and stop pre-zeroing counter by [@yuneng-berri](https://github.com/yuneng-berri) in [#29358](https://github.com/BerriAI/litellm/pull/29358) - test(e2e): cover PROXY\_LOGOUT\_URL redirect on Logout by [@ryan-crabbe-berri](https://github.com/ryan-crabbe-berri) in [#29080](https://github.com/BerriAI/litellm/pull/29080) - fix(ui): break logout redirect loop across dev and proxy origins by [@yuneng-berri](https://github.com/yuneng-berri) in [#29360](https://github.com/BerriAI/litellm/pull/29360) - fix(openai-moderation): wire streaming flags through to unified dispatcher by [@michelligabriele](https://github.com/michelligabriele) in [#27324](https://github.com/BerriAI/litellm/pull/27324) - chore(ci): build ui by [@yuneng-berri](https://github.com/yuneng-berri) in [#29366](https://github.com/BerriAI/litellm/pull/29366) - fix(v3 limiter): cap no-max\_tokens TPM floor at smallest configured limit by [@michelligabriele](https://github.com/michelligabriele) in [#28805](https://github.com/BerriAI/litellm/pull/28805) - fix(e2e): tolerate trailing slash in SERVER\_ROOT\_PATH login redirect by [@yuneng-berri](https://github.com/yuneng-berri) in [#29369](https://github.com/BerriAI/litellm/pull/29369) - chore(deps): bump deps by [@yuneng-berri](https://github.com/yuneng-berri) in [#29373](https://github.com/BerriAI/litellm/pull/29373) - chore(ci): promote internal staging to main by [@yuneng-berri](https://github.com/yuneng-berri) in [#29372](https://github.com/BerriAI/litellm/pull/29372) - chore(release): patch v1.88.0-rc.1 with four staged fixes by [@mateo-berri](https://github.com/mateo-berri) in [#29632](https://github.com/BerriAI/litellm/pull/29632) - chore(release): patch v1.88.0-rc.1 with [#29612](https://github.com/BerriAI/litellm/issues/29612) (session-token budget-ceiling exemption) by [@mateo-berri](https://github.com/mateo-berri) in [#29637](https://github.com/BerriAI/litellm/pull/29637) - fix(key\_generate): harden GHSA-q775 …

tin-berri force-pushed the litellm_mcp_tools_obomode branch 2 times, most recently from b190ba3 to 7eb31bb Compare June 5, 2026 23:21

veria-ai Bot reviewed Jun 5, 2026

View reviewed changes

tin-berri force-pushed the litellm_mcp_tools_obomode branch from 7eb31bb to 63c47c5 Compare June 5, 2026 23:56

updated logic to store access_tokens and refresh_tokens in DB

bf8652a

tin-berri force-pushed the litellm_mcp_tools_obomode branch from 63c47c5 to bf8652a Compare June 6, 2026 00:13

tin-berri added 2 commits June 5, 2026 17:29

fix(mcp): refresh expired OBO OAuth tokens on the Tools-tab list path

7a29f2c

Merge remote-tracking branch 'origin/litellm_internal_staging' into l…

623a14b

…itellm_mcp_tools_obomode # Conflicts: # tests/test_litellm/proxy/_experimental/mcp_server/test_db_credentials.py

tin-berri self-assigned this Jun 6, 2026

tin-berri requested a review from mateo-berri June 6, 2026 04:36

mateo-berri approved these changes Jun 6, 2026

View reviewed changes

mateo-berri merged commit 22186f4 into litellm_internal_staging Jun 6, 2026
118 checks passed

mateo-berri deleted the litellm_mcp_tools_obomode branch June 6, 2026 05:29

tin-berri changed the title ~~fix(ui): persist Tools-tab MCP OAuth token to DB~~ fix(MCP): persist Tools-tab MCP OAuth token to DB Jun 8, 2026

tin-berri changed the title ~~fix(MCP): persist Tools-tab MCP OAuth token to DB~~ fix(mcp): persist Tools-tab MCP OAuth token to DB Jun 8, 2026

tin-berri mentioned this pull request Jun 10, 2026

Adhoc v1.88.0 backport: MCP OAuth/Tools fixes + allowlist update-500 fix (#29411) + card grid #30136

Closed

6 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

fix(mcp): persist Tools-tab MCP OAuth token to DB#29809

fix(mcp): persist Tools-tab MCP OAuth token to DB#29809
mateo-berri merged 3 commits into
litellm_internal_stagingfrom
litellm_mcp_tools_obomode

tin-berri commented Jun 5, 2026 •

edited

Loading

Uh oh!

greptile-apps Bot commented Jun 5, 2026

Important Files Changed

Uh oh!

codecov Bot commented Jun 5, 2026 •

edited

Loading

Uh oh!

veria-ai Bot Jun 5, 2026

Uh oh!

tin-berri Jun 6, 2026

Uh oh!

veria-ai Bot commented Jun 5, 2026

Uh oh!

mateo-berri left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

tin-berri commented Jun 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Problem

Fix

Linear ticket

Note

Pre-Submission checklist

CI (LiteLLM team)

Screenshots / Proof of Fix

Token refresh on expiry (the refresh half)

Type

Changes

Uh oh!

greptile-apps Bot commented Jun 5, 2026

Greptile Summary

Confidence Score: 4/5

Important Files Changed

Uh oh!

codecov Bot commented Jun 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

veria-ai Bot Jun 5, 2026

Choose a reason for hiding this comment

Uh oh!

tin-berri Jun 6, 2026

Choose a reason for hiding this comment

Uh oh!

veria-ai Bot commented Jun 5, 2026

PR overview

Open issues (1)

Uh oh!

mateo-berri left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

tin-berri commented Jun 5, 2026 •

edited

Loading

codecov Bot commented Jun 5, 2026 •

edited

Loading