Feature/upgrade to v1.81.0 stable by shriharsha98 · Pull Request #52 · juspay/litellm

shriharsha98 · 2026-02-05T09:54:43Z

Title

Relevant issues

Pre-Submission checklist

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

I have Added testing in the tests/litellm/ directory, Adding at least 1 test is a hard requirement - see details
I have added a screenshot of my new test passing locally
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

Type

🆕 New Feature
🐛 Bug Fix
🧹 Refactoring
📖 Documentation
🚄 Infrastructure
✅ Test

Changes

…mpletions [fix] mcp chat completions

…e_gen Add openrouter support for image/generation endpoints

…BerriAI#19136)

allows rendering on models.litellm.ai

… from docs maintain just 1 place for tutorials

…bedding_header_forwarding_with_model_group

…put_array_of_tokens

easier to surface

…ests Stabilise mock tests

…ime_fix Fix model name from query param in realtime request

…19148) * Added ability to customize logfire base url through env var * Added test to check if env var is used correctly for logfire * Document the env var * Documented env var in config_settings.md

[Docs] Deleted Keys and Teams Docs

…violation-msg feat(panw_prisma_airs): add custom violation message support

[Infra] Fix UI Build

… Code (BerriAI#19294) * fix websearch_interception_converted_stream * test_websearch_interception_no_tool_call_streaming * FakeAnthropicMessagesStreamIterator * LITELLM_WEB_SEARCH_TOOL_NAME * fixes tools def for litellm web search * fixes FakeAnthropicMessagesStreamIterator * test_litellm_standard_websearch_tool * use new hook for modfying before any transfroms from litellm * init WebSearchInterceptionLogger + ARCHITECTURE * fix config.yaml * init doc for claude code web search * docs fix * doc fix * fix mypy linting

…orwarded down to target (BerriAI#19524) * test_vertex_passthrough_forwards_anthropic_beta_header * add_incoming_headers

…ed down to LLM API (BerriAI#19542) * fix ALLOWED_VERTEX_AI_PASSTHROUGH_HEADERS * test_vertex_passthrough_forwards_anthropic_beta_header * fix test_vertex_passthrough_forwards_anthropic_beta_header * test_vertex_passthrough_does_not_forward_litellm_auth_token * fix utils * Using Anthropic Beta Features on Vertex AI * test_forward_headers_from_request_x_pass_prefix

…erriAI#19424)

…n a separate process (BerriAI#19427)

patches preserved: - FREE_MODELS budget bypass system (auth_checks.py, db_spend_update_writer.py, proxy_track_cost_callback.py) - Vision model routing with fallback (router.py, litellm_pre_call_utils.py, _types.py) - DAU/WAU/MAU analytics with team filtering (networking.tsx, new_usage.tsx) - GCS bucket logging integration (integrations/gcs_bucket/) - Budget duration support - daily/weekly/monthly (schema.prisma fields: budget_duration, budget_reset_at) Upstream improvements integrated: - Organization and team member budget checks (auth_checks.py) - Security fixes (CVE patches, expired key leak prevention) - PostgreSQL cached plan fixes - SIGTERM handling improvements - Budget validation enhancements - MCP server management mode controls - Extended EntityType support (organization, customer, agent) - Enhanced entity usage export (CSV/JSON utilities) Conflict resolution summary: - 104 total conflicts resolved (19 code files, 85 UI build artifacts) - Critical backend files auto-merged successfully (schema.prisma, auth_checks.py, router.py, litellm_pre_call_utils.py, proxy_track_cost_callback.py) - Manual resolutions: _types.py, db_spend_update_writer.py, GCS Readme, UI TypeScript files - UI build artifacts accepted from upstream (will be regenerated on next build) Testing required: - FREE_MODELS bypass functionality - Vision model fallback routing - Budget duration enforcement (daily/weekly/monthly) - DAU/WAU/MAU analytics endpoints - Team-based filtering - GCS logging to buckets - Organization budget checks - Team member budget checks

Copilot

Pull request overview

This PR upgrades LiteLLM to v1.81.0 stable, introducing significant new provider support, API endpoints, and infrastructure improvements. The update adds 4 new LLM providers, 2 new API endpoints, extensive documentation updates, and enhanced deployment configurations.

Changes:

Added support for 4 new providers (Apertis AI, Abliteration, Amazon Nova, and enhanced Azure features)
Introduced 2 new API endpoints (/interactions and /v1/messages/count_tokens)
Enhanced documentation with 100+ new guides, provider docs, and integration examples
Improved Docker and Helm deployment configurations with security hardening
Updated CI/CD pipelines and testing infrastructure

Reviewed changes

Copilot reviewed 140 out of 2393 changed files in this pull request and generated 6 comments.

Show a summary per file

File	Description
docs/my-website/docs/providers/*	Added documentation for new providers (Apertis, Abliteration, Amazon Nova, Azure enhancements)
docs/my-website/docs/observability/*	Added integration docs for Qualifire, Levo AI, Focus, Generic API, updated Phoenix and Datadog
docs/my-website/docs/projects/*	Added documentation for OpenAI Agents SDK, Harbor, GraphRAG, Google ADK, Agent Lightning
docs/my-website/docs/*.md	Added new endpoint documentation for /interactions, /anthropic_count_tokens, code interpreter, agent permissions
docker/*	Updated Dockerfiles with Windows line ending conversion and security improvements
deploy/charts/litellm-helm/*	Enhanced Helm chart with ServiceMonitor, topology spread, lifecycle hooks, and version updates
.github/workflows/*	Updated CI workflows with new labeling, duplicate checking, and improved testing
ci_cd/*	Enhanced security scanning with Grype config and secret detection preparation
cookbook/*	Added Braintrust prompt wrapper and AI coding tool guides

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-02-05T09:56:02Z

docs/my-website/docs/oidc.md

      aws_session_name: "my-test-session"
      aws_role_name: "arn:aws:iam::335785316107:role/litellm-github-unit-tests-circleci"
-      aws_web_identity_token: "oidc/circleci_v2/"
+      aws_web_identity_token: "oidc/example-provider/"


The line changed from 'circleci_v2/' to 'example-provider/', but the section header still refers to CircleCI. Either update line 112's header to match the generic example, or revert this change to use a CircleCI-specific path.

Suggested change

aws_web_identity_token: "oidc/example-provider/"

aws_web_identity_token: "oidc/circleci_v2/"

Copilot · 2026-02-05T09:56:03Z

docs/my-website/docs/providers/bedrock_agentcore.md


+:::info
+
+This documentation is for **AgentCore Agents** (agent runtimes). If you want to use AgentCore MCP servers, add them as you would any other MCP server. See the [MCP documentation](https://docs.litellm.ai/docs/mcp) for details.


The MCP documentation link should be relative using '../mcp' instead of the full path to maintain consistency with other documentation links and avoid breakage if the domain changes.

Suggested change

This documentation is for **AgentCore Agents** (agent runtimes). If you want to use AgentCore MCP servers, add them as you would any other MCP server. See the [MCP documentation](https://docs.litellm.ai/docs/mcp) for details.

This documentation is for **AgentCore Agents** (agent runtimes). If you want to use AgentCore MCP servers, add them as you would any other MCP server. See the [MCP documentation](../mcp) for details.

Copilot · 2026-02-05T09:56:03Z

docker/supervisord.conf

@@ -14,6 +14,7 @@ priority=1
 exitcodes=0
 stopasgroup=true
 killasgroup=true


The new SUPERVISORD_STOPWAITSECS environment variable is referenced but not documented. Add a comment explaining its purpose and default value, or document it in the deployment guide.

Suggested change

killasgroup=true

killasgroup=true

; SUPERVISORD_STOPWAITSECS controls how many seconds supervisord waits for a

; graceful shutdown before sending SIGKILL. If not set, supervisord's default

; stopwaitsecs (10 seconds) applies. This value is shared by both main and health.

Copilot · 2026-02-05T09:56:03Z

docker/supervisord.conf

 exitcodes=0
 stopasgroup=true
 killasgroup=true
+stopwaitsecs=%(ENV_SUPERVISORD_STOPWAITSECS)s


The new SUPERVISORD_STOPWAITSECS environment variable is referenced but not documented. Add a comment explaining its purpose and default value, or document it in the deployment guide.

Copilot · 2026-02-05T09:56:03Z

docker/prod_entrypoint.sh

@@ -2,6 +2,7 @@

 if [ "$SEPARATE_HEALTH_APP" = "1" ]; then
    export LITELLM_ARGS="$@"


The default value of 3600 seconds (1 hour) should be documented. Consider adding a comment explaining why this long timeout is necessary for graceful shutdown.

Suggested change

export LITELLM_ARGS="$@"

export LITELLM_ARGS="$@"

# Use a long default (1 hour) so supervisord has enough time to gracefully stop

# long-running worker processes and in-flight requests during shutdown.

Copilot · 2026-02-05T09:56:04Z

deploy/charts/litellm-helm/values.yaml

@@ -3,6 +3,7 @@
 # Declare variables to be passed into your templates.

 replicaCount: 1


The new numWorkers parameter is commented out but lacks documentation. Add a comment explaining what it controls and when users should configure it.

Suggested change

replicaCount: 1

replicaCount: 1

# Number of worker processes per pod for handling LiteLLM requests; adjust this on larger nodes or under high concurrency, or leave unset to use the default.

…/upgrade-to-v1.81.0-stable

uc4w6c and others added 30 commits January 15, 2026 16:04

fix: lint

0db6229

fix: prevent MCP type objects from being captured in locals()

9d173ba

Merge pull request BerriAI#19129 from BerriAI/litellm_fix_mcp_chat_co…

f541bc0

…mpletions [fix] mcp chat completions

Fix: BerriAI#19089 websocket version error

a3cf178

Fix: handling of model name in query param

e9a2999

Add logger to see websocket request

7f79140

Fix: code quality tests

58eef29

Fix: code quality tests

2023982

Merge pull request BerriAI#19059 from BerriAI/litellm_openrouter_imag…

83cdfd8

…e_gen Add openrouter support for image/generation endpoints

docs(logging.md): add guide for mounting custom callbacks in Helm/K8s (…

8804c78

…BerriAI#19136)

docs: new cookbook .md for claude code

7c53bf1

allows rendering on models.litellm.ai

refactor: rename file for consistency

02d782f

Add user auth in standard logging object for bedrock passthrough

eb49adb

remove unused space

3ebf16e

Fix: test router

3af994a

fix(index.json): have a parent index.json and just link out to guides…

37abfc6

… from docs maintain just 1 place for tutorials

Fix: tests/test_litellm/proxy/test_litellm_pre_call_utils.py::test_em…

f28d951

…bedding_header_forwarding_with_model_group

Fix: tests/test_litellm/proxy/test_proxy_server.py::test_embedding_in…

4bdda9c

…put_array_of_tokens

Fix: response enterprise tests

b4ee1b2

docs(claude_mcp.md): separate claude mcp tutorial into a separate doc

c7ca2dd

easier to surface

docs: update index.json

04eec6c

fix: mount config.yaml as single file in Helm chart (BerriAI#19146)

3ad8fa5

Fix: response enterprise tests

890fa85

Fix: mock test tests

83e3394

Fix: mock test tests

8c454eb

Merge pull request BerriAI#19141 from BerriAI/litellm_stablise_mock_t…

0ae471a

…ests Stabilise mock tests

Merge pull request BerriAI#19135 from BerriAI/litellm_websocker_realt…

8b7c36c

…ime_fix Fix model name from query param in realtime request

Adjust icons for buttons

79d35ec

fixing build

d44b147

Added ability to customize logfire base url through env var (BerriAI#…

c619569

…19148) * Added ability to customize logfire base url through env var * Added test to check if env var is used correctly for logfire * Document the env var * Documented env var in config_settings.md

yuneng-jiang and others added 25 commits January 17, 2026 13:19

deleted keys and teams docs

19a69a8

Merge pull request BerriAI#19291 from BerriAI/deleted_keys_docs_2

953e273

[Docs] Deleted Keys and Teams Docs

Merge pull request BerriAI#19272 from Harshit28j/feature/panw-custom-…

d28bf98

…violation-msg feat(panw_prisma_airs): add custom violation message support

fix build attempt

c1f194c

testing adding entire out

91eb047

Merge pull request BerriAI#19293 from BerriAI/litellm_ui_build_fix_22

42c0136

[Infra] Fix UI Build

test_router_fallbacks_with_custom_model_costs

5812654

test_deepseek_mock_completion

db7de13

v1.81.0

0a84120

docs fix

c30b17a

test_aiohttp_openai

60dd04a

fix

e15526a

qa fixes

7d24bbe

docs fix

4610d1d

docs fix

c158f83

docs fix

c699882

docs fix

26497b4

docs fix

534fa9f

[Fix] LiteLLM VertexAI Pass through - ensuring incoming headers are f…

ad1db9f

…orwarded down to target (BerriAI#19524) * test_vertex_passthrough_forwards_anthropic_beta_header * add_incoming_headers

Fix: Handle PostgreSQL cached plan errors during rolling deployments (B…

9e893b7

…erriAI#19424)

Fix in-flight request termination on SIGTERM when health-check runs i…

790a5ce

…n a separate process (BerriAI#19427)

Fix build errors after merge

6d3721f

Copilot AI review requested due to automatic review settings February 5, 2026 09:54

Copilot AI reviewed Feb 5, 2026

View reviewed changes

Merge branch 'release/prod' of github.com:juspay/litellm into feature…

9cef052

…/upgrade-to-v1.81.0-stable

prakhar-prakash-juspay approved these changes Feb 10, 2026

View reviewed changes

shriharsha98 merged commit ed2b8e9 into release/prod Feb 10, 2026
1 check passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Feature/upgrade to v1.81.0 stable#52

Feature/upgrade to v1.81.0 stable#52
shriharsha98 merged 2862 commits intorelease/prodfrom
feature/upgrade-to-v1.81.0-stable

shriharsha98 commented Feb 5, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Feb 5, 2026

Uh oh!

Copilot AI Feb 5, 2026

Uh oh!

Copilot AI Feb 5, 2026

Uh oh!

Copilot AI Feb 5, 2026

Uh oh!

Copilot AI Feb 5, 2026

Uh oh!

Copilot AI Feb 5, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

20 participants

	aws_web_identity_token: "oidc/example-provider/"
	aws_web_identity_token: "oidc/circleci_v2/"


		:::info

		This documentation is for AgentCore Agents (agent runtimes). If you want to use AgentCore MCP servers, add them as you would any other MCP server. See the [MCP documentation](https://docs.litellm.ai/docs/mcp) for details.

-killasgroup=true
+killasgroup=true
+; SUPERVISORD_STOPWAITSECS controls how many seconds supervisord waits for a
+; graceful shutdown before sending SIGKILL. If not set, supervisord's default
+; stopwaitsecs (10 seconds) applies. This value is shared by both main and health.

		@@ -2,6 +2,7 @@

		if [ "$SEPARATE_HEALTH_APP" = "1" ]; then
		export LITELLM_ARGS="$@"

		@@ -3,6 +3,7 @@
		# Declare variables to be passed into your templates.

		replicaCount: 1

	replicaCount: 1
	replicaCount: 1
	# Number of worker processes per pod for handling LiteLLM requests; adjust this on larger nodes or under high concurrency, or leave unset to use the default.

Conversation

shriharsha98 commented Feb 5, 2026

Title

Relevant issues

Pre-Submission checklist

Type

Changes

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Feb 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 5, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

20 participants