scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-10-10 03:28:53 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 86
39566 commits 3393 branches 1449 tags 5.1 GiB
Commit graph

39566 commits

This branch
This branch
All branches
Author SHA1 Message Date
joereyna
180e72d15d fix: add oauth2_flow=client_credentials to MCPServer test helper 2026-03-12 12:22:18 -07:00
Cesar Garcia
ec763784e0
Merge branch 'main' into litellm_oss_staging_03_11_2026 2026-03-12 16:21:28 -03:00
Joe Reyna
f2f843448e
Merge pull request #23414 from joereyna/fix/pass-through-server-root-path 
fix: strip SERVER_ROOT_PATH prefix before checking mapped pass-through routes
 2026-03-12 11:57:13 -07:00
yuneng-jiang
779a231036
Merge pull request #23465 from BerriAI/ui_build_mar12 
[Infra[ UI Build + Tests
 2026-03-12 11:29:46 -07:00
yuneng-jiang
d17d205963 chore: update Next.js build artifacts (2026-03-12 18:25 UTC, node v22.16.0) 2026-03-12 11:25:05 -07:00
yuneng-jiang
c8ce098b65 fixing tests 2026-03-12 11:24:15 -07:00
yuneng-jiang
68b8b47d58
Merge pull request #23463 from BerriAI/ui_build_mar12 
[Infra] Fixing UI Build
 2026-03-12 10:54:00 -07:00
yuneng-jiang
e38edb6190 fixing ui build 2026-03-12 10:51:49 -07:00
Chesars
0fc407cfdd ci: exclude enterprise/ from black --check in linting workflow 
Contributors don't have local access to enterprise/ files,
so the check would always fail on unformatted enterprise code.
 2026-03-12 14:27:00 -03:00
Cesar Garcia
f79744cee2
Merge pull request #18648 from Chesars/fix-black-check-ci 
fix: check Black formatting in CI instead of auto-formatting
 2026-03-12 14:24:37 -03:00
Chesars
0fcc36c301 style: run black formatter on 52 non-enterprise files 
Formats all files flagged by CI except 12 enterprise/ files
which require separate access to format.
 2026-03-12 14:23:50 -03:00
Sameer Kankute
3c322a879f
Merge pull request #23460 from BerriAI/litellm_add_webrtc_support 
Improve doc for WebRTC
 2026-03-12 22:46:23 +05:30
Sameer Kankute
8bbebb5d75 Improve doc for WebRTC 2026-03-12 22:45:36 +05:30
Sameer Kankute
5ccf1cf4fe
Merge pull request #23459 from BerriAI/litellm_add_webrtc_support 
Improve doc for WebRTC
 2026-03-12 22:43:05 +05:30
Sameer Kankute
cf1d330e63 Improve doc for WebRTC 2026-03-12 22:41:39 +05:30
Cesar Garcia
e01d722803
Merge branch 'main' into litellm_oss_staging_03_11_2026 2026-03-12 13:53:14 -03:00
Sameer Kankute
d507f840d3
Merge pull request #23432 from BerriAI/litellm_azure-model-router-show-actual-model 
feat(azure_ai): show actual model used in Azure Model Router response
 2026-03-12 22:18:50 +05:30
Sameer Kankute
d1a99f571e
Merge pull request #23446 from BerriAI/litellm_add_webrtc_support 
[Feat] Add WebRTC support
 2026-03-12 22:16:55 +05:30
Sameer Kankute
507bced04f Fix the live tester 2026-03-12 22:14:08 +05:30
Cesar Garcia
33457ab702
Merge pull request #23457 from Chesars/revert-revert-21601-model-cost-aliases 
fix: restore commits lost by revert of PR #21601
 2026-03-12 13:43:07 -03:00
Chesars
4e6e1d8de8 merge: resolve conflicts with upstream staging (bedrock + mcp tests) 
Keep both sets of tests: upstream's OAuth2 token injection test and
our case-insensitive tool matching tests. Use upstream's version of
the bedrock output_config test (more comprehensive).
 2026-03-12 13:40:16 -03:00
Chesars
feed274aa3 Reapply "feat: add model_cost aliases expansion support" 
This reverts commit 3d2df7e8b5.
 2026-03-12 13:36:57 -03:00
michelligabriele
7c5e2e8389
fix(proxy): make async_post_call_response_headers_hook consistent across all endpoints (#22985) 
* fix(proxy): make async_post_call_response_headers_hook consistent across all endpoints

The response headers hook had 5 gaps that prevented callbacks from
reliably extracting routing metadata across endpoint types:

1. Hook never fired for /audio/transcriptions (endpoint bypasses
   base_process_llm_request)
2. custom_llm_provider not accessible in hook data for any endpoint
3. custom_llm_provider not stamped in ResponsesAPIResponse._hidden_params
   (unlike chat completions)
4. model_info under inconsistent keys (metadata vs litellm_metadata)
5. request_headers always None at all call sites

This adds a litellm_call_info parameter to the hook that normalizes
routing metadata (custom_llm_provider, model_info, api_base, model_id)
regardless of endpoint type. Also stamps custom_llm_provider on
Responses API responses, adds the hook call to the transcription
handler, and passes request_headers at all call sites.

Supersedes PR #21385.

* fix(proxy): address review feedback — safer backwards compat and None guards

- Replace try/except TypeError with inspect.signature() check for
  litellm_call_info backwards compatibility. This avoids masking real
  TypeErrors inside callback implementations and prevents double
  invocation with inconsistent parameters.

- Use (data.get("key") or {}) instead of data.get("key", {}) to guard
  against keys that exist with an explicit None value, which would
  cause AttributeError on the subsequent .get() call.

* fix(proxy): cache inspect.signature result for callback compat check

Move the inspect.signature() call into a module-level helper with a
dict cache keyed by callback identity. Avoids repeated introspection
per request per callback in the hot path.

* fix(proxy): use class identity for signature cache key

Key the _CALLBACK_ACCEPTS_CALL_INFO cache by id(type(cb)) instead of
id(cb) to avoid stale entries from Python address reuse after GC.
All instances of the same callback class share the same method
signature, so class identity is both safer and more cache-efficient.
 2026-03-12 08:51:00 -07:00
joereyna
1af7f11dae fix: extract normalize_route_for_root_path to deduplicate root-path stripping; fix mock target 2026-03-12 08:16:00 -07:00
joereyna
938452cc59 fix: extract normalize_route_for_root_path to deduplicate root-path stripping 2026-03-12 07:55:22 -07:00
Cesar Garcia
6bd7cd7573
Merge branch 'main' into litellm_oss_staging_03_11_2026 2026-03-12 10:43:08 -03:00
Sameer Kankute
291e6e1841
Merge pull request #23435 from BerriAI/litellm_vector-store-retrieve-list-update-delete 
Add vector store retrieve list update delete
 2026-03-12 19:08:39 +05:30
Sameer Kankute
4f5b6ae556
Merge pull request #23448 from BerriAI/litellm_cicd_1203126 
Litellm cicd 1203126
 2026-03-12 19:07:33 +05:30
Sameer Kankute
b7cfcdd35d Add docs 2026-03-12 19:06:57 +05:30
Chesars
fa68d69bcf fix: restore _get_effort_level and is_model_gpt_5_4_plus_model (PR #23151) 
Independent fix (base: main) collaterally removed by PR #23276.
Restores:
- _get_effort_level() for extracting effort from string or dict
- is_model_gpt_5_4_plus_model() classmethod
- effective_effort usage in xhigh/tool-drop/sampling/temperature guards
- Azure: _get_effort_level import and usage for dict reasoning_effort
- Azure: gpt-5.4+ tool+reasoning drop logic
 2026-03-12 10:28:27 -03:00
Chesars
9d1bcc7d37 Merge upstream/main into litellm_oss_staging_03_11_2026 
Restore independent fixes from main that were collaterally removed
when PR #23276 (staging_03_10 → main) carried a revert commit:
- bedrock: restore output_config pop (PR #23240)
- redact_messages: restore dict handling for ModelResponse (PR #23235)
- model_checks: restore list() copies to avoid cache mutation (PR #23236)
- openapi_to_mcp_generator: restore relative URL handling (PR #23238)
- vertex_ai/gemini: restore _LITELLM_INTERNAL_EXTRA_BODY_KEYS check (PR #23131)
- openai types: restore extra finish reasons (PR #22138)
- completion_extras: restore usage transformation logic

Accept main for: model_prices JSONs, credential_endpoints,
team_endpoints, object_permission_utils, responses transformation.
 2026-03-12 10:25:13 -03:00
Sameer Kankute
bb451cfcb0 address greptile review feedback (greploop iteration 2) 
- Thread api_version through HTTP handlers to Azure realtime endpoints
- Make expires_at optional in RealtimeClientSecretResponse
- Fix test token expiry times to be in the future
- Populate user_id and team_id in minimal_auth for spend tracking

Made-with: Cursor
 2026-03-12 18:53:22 +05:30
Sameer Kankute
f4103c51a6 address greptile review feedback (greploop iteration 1) 
- Add api-version query param to Azure realtime URLs
- Remove Content-Type from Azure realtime_calls headers (httpx sets it)
- Add token expiry validation in proxy_realtime_calls endpoint
- Fix type annotations for upstream_resp

Made-with: Cursor
 2026-03-12 18:40:37 +05:30
Sameer Kankute
f5be79419c Fix test_claude_agent_sdk_streaming 2026-03-12 18:36:04 +05:30
Sameer Kankute
f6238e781e Fix mypy 2026-03-12 18:35:17 +05:30
Sameer Kankute
982f3917c5 Fix test_standard_logging_payload 2026-03-12 18:35:01 +05:30
Sameer Kankute
15d873e204 Fix update deprecated model test 2026-03-12 18:34:20 +05:30
Sameer Kankute
374c35a6b7 Fix update deprecated model test 2026-03-12 18:34:15 +05:30
Sameer Kankute
7c70015a5f Fix mcp error 2026-03-12 18:33:58 +05:30
Sameer Kankute
18df137021 Fix mypy error 2026-03-12 18:33:40 +05:30
Sameer Kankute
0f91a4f9da Fix test_get_tools_for_single_server 2026-03-12 18:33:14 +05:30
Sameer Kankute
412a283569 Revert "fix(vertex): skip harmful schema transforms for Gemini 2.0+ tool parameters" 
This reverts commit a9c3095cc5.
 2026-03-12 18:26:11 +05:30
Sameer Kankute
72c98489d1 Revert "fix(vertex): shallow copy parameters before mutating in _build_vertex_schema_for_gemini_2" 
This reverts commit 08d81f5d7c.
 2026-03-12 18:26:11 +05:30
Chesars
47e9a0dcf0 Merge branch 'litellm_oss_staging_03_11_2026' of https://github.com/BerriAI/litellm into litellm_oss_staging_03_11_2026 2026-03-12 09:39:22 -03:00
Chesars
1be6b31e2f merge: resolve conflicts between main and litellm_oss_staging_03_11_2026 2026-03-12 09:38:31 -03:00
Sameer Kankute
7aa5bd3ff3
Merge pull request #23429 from BerriAI/litellm_dev_03_10_2026_p1 
Litellm dev 03 10 2026 p1
 2026-03-12 18:04:48 +05:30
Sameer Kankute
7778af6c78 Add tests 2026-03-12 17:54:57 +05:30
Sameer Kankute
e2be1aabae Add webrtc in init 2026-03-12 16:37:36 +05:30
Sameer Kankute
eb64cd6c46 Add webrtc transformations and http handler 2026-03-12 16:37:23 +05:30
Sameer Kankute
f793d2043b Add webrtc routing 2026-03-12 16:36:59 +05:30