BerriAI/litellm - CodSpeed

litellm

Blog Docs Changelog

Performance History

Latest Results

test(passthrough): intercept AsyncClient.send in legacy passthrough tests and assert final wire params

litellm_stream_passthrough_get_responses

35 seconds ago

fix(passthrough): stop request params from clobbering merged target query params

litellm_fix_merge_query_params_clobber

2 minutes ago

Remove dashboard brief from e2e coverage PR

litellm_e2e_llm_module_breakdown

20 minutes ago

Add e2e coverage dashboard metrics

ishaan-berri:litellm_e2e_llm_module_breakdown

40 minutes ago

fix(mcp): bind oauth_delegate discovery resource to the upstream oauth_delegate forwards the caller's token to the upstream, which validates its audience, so the protected-resource metadata must keep resource pointing at the upstream (returned verbatim, like true_passthrough) rather than rewriting it to the gateway. Rewriting to the gateway asks the client to mint a token bound to the gateway audience, which a strict IdP (Entra) refuses to issue for an unregistered resource and a spec-compliant upstream rejects on receipt. The legacy is_oauth_passthrough opt-in keeps the gateway rewrite unchanged.

litellm_mcp_passthrough_delegate_modes

46 minutes ago

fix(ui): enforce exclusive selection of proxy/team sentinels in key edit form The edit form models Select had a pass-through onChange, so a user could submit a mixed list like ["all-proxy-models", "gpt-4"] even though the sentinels are meant to be exclusive. Mirror the create form and collapse the selection to the sentinel when all-proxy-models or all-team-models is picked; add a regression test that selects a model plus the proxy sentinel and asserts the submitted models list is just ["all-proxy-models"] Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

litellm_key_models_dropdown_team_gating

47 minutes ago

test(mcp): pin cache invalidation on the revoke already-gone branch Greptile's review flagged that only the happy-path delete asserted the invalidate; a refactor moving the call inside the try block would silently skip the cache drop when the row was already deleted by a concurrent request while the cache still held the revoked token. The new test fails on exactly that mutation

litellm_mcp_oauth_cache_invalidate_on_reauth

57 minutes ago

fix(mcp): accept integer progressToken in host progress capture

litellm_mcp_int_progress_token

57 minutes ago

Latest Branches

0%

fix(passthrough): stream non-sse passthrough responses instead of buffering in memory#32386

21 minutes ago

3107c8b

litellm_stream_passthrough_get_responses

0%

fix(passthrough): stop request params from clobbering merged target query params#32404

25 minutes ago

dd989af

litellm_fix_merge_query_params_clobber

0%

Split LLM e2e coverage modules#32403

41 minutes ago

52033ef

litellm_e2e_llm_module_breakdown

© 2026 CodSpeed Technology

Home Terms Privacy Docs