scriptoriumadmin
  • Joined on 2026-03-01
scriptoriumadmin synced commits to refs/pull/48948/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
8ff495be96 fix(proxy): price OpenAI pass-through /v1/decisions calls like native Decisions (#49392)
be4f3d36ce feat(accounting): capture request usage through the durable meter (#49133)
a713ed6bab fix(gemini): add thought_signature_in_tool_call_id to keep tool call ids short for clients that cap them (#47692)
febd340120 fix(guardrails)!: keep the upstream error when a /v1/messages stream ends in one (#49088)
Compare 43 commits »
scriptoriumadmin synced commits to refs/pull/48936/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
c43a1addf2 refactor(router): extract PythonCore behind RouterCore and pass the router its config (#49119)
a5f980600c fix(a2a): answer proxy refusals on /a2a POST routes as JSON-RPC errors (#49476)
7143aeeb86 feat(passthrough): add Deepgram streaming text-to-speech WebSocket route (#49282)
3129e44848 ci: cut the weekly rc branch on Monday 3am PT instead of Friday (#49472)
Compare 88 commits »
scriptoriumadmin synced commits to refs/pull/48932/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
8ff495be96 fix(proxy): price OpenAI pass-through /v1/decisions calls like native Decisions (#49392)
be4f3d36ce feat(accounting): capture request usage through the durable meter (#49133)
a713ed6bab fix(gemini): add thought_signature_in_tool_call_id to keep tool call ids short for clients that cap them (#47692)
febd340120 fix(guardrails)!: keep the upstream error when a /v1/messages stream ends in one (#49088)
Compare 24 commits »
scriptoriumadmin synced commits to refs/pull/48942/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
c43a1addf2 refactor(router): extract PythonCore behind RouterCore and pass the router its config (#49119)
a5f980600c fix(a2a): answer proxy refusals on /a2a POST routes as JSON-RPC errors (#49476)
7143aeeb86 feat(passthrough): add Deepgram streaming text-to-speech WebSocket route (#49282)
3129e44848 ci: cut the weekly rc branch on Monday 3am PT instead of Friday (#49472)
Compare 46 commits »
scriptoriumadmin synced commits to refs/pull/48946/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
cee72ffdd6 fix(proxy): count native Responses WebSocket turn tokens toward key and tag TPM limits (#49257)
da71a03f6e fix(proxy): trace model list requests with the caller's traceparent (#49141)
4cbd830315 fix(guardrails): keep a pipeline step's guardrail list out of litellm_metadata, name Lakera v1 on a block, inject Lakera v2's client
c4b01b2ac8 fix(decisions)!: honor a deployment's own supports_audio_input and supports_video_input flags (#49207)
Compare 81 commits »
scriptoriumadmin synced commits to refs/pull/48958/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
60fc121ce8 test(a2a): replace the third-party agent e2e test with an integration test (#49541)
1b3da63703 test: move Sail, Weave and Presidio e2e tests to integration tests (#49533)
187f936cb3 fix(proxy)!: support reload with multiple workers (#45909)
a1c5829f6b fix(bedrock): raise GovCloud West Llama 3 max output to the model card 8K (#49543)
Compare 124 commits »
scriptoriumadmin synced commits to refs/pull/48967/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
6aa546303e docs(pr-template): wrap the agent-drafted summary in make-me-human-readable tags (#49494)
580b9266a1 feat(chat): resume replies after browser disconnects (#49461)
d2321581ec fix(cohere): emit streamed tool calls from v2 tool-call-start and tool-call-delta chunks (#47794)
f7b2cd0000 fix(anthropic): answer 400 naming a non-string role or tool call id instead of 500 (#49237)
Compare 79 commits »
scriptoriumadmin synced commits to refs/pull/48979/head at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
f414a42ecf fix(passthrough): keep the session update a buffered clip ran under until its answer is billed
3e8aefd633 Merge remote-tracking branch 'origin/main' into litellm_openai_ws_frame_model_gate
c43a1addf2 refactor(router): extract PythonCore behind RouterCore and pass the router its config (#49119)
a5f980600c fix(a2a): answer proxy refusals on /a2a POST routes as JSON-RPC errors (#49476)
7143aeeb86 feat(passthrough): add Deepgram streaming text-to-speech WebSocket route (#49282)
Compare 46 commits »
scriptoriumadmin synced commits to refs/pull/48979/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
f414a42ecf fix(passthrough): keep the session update a buffered clip ran under until its answer is billed
ddab201d68 test(translation): expect the EOF error frame when an Anthropic json_schema stream is cut (#49521)
22d0b1e443 fix(router): restore four base behaviors the RoutingContext refactor changed (#49395)
a7425101bb fix(a2a): log authenticated pre-dispatch agent rejections (#49492)
Compare 56 commits »
scriptoriumadmin synced commits to refs/pull/48997/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
3cb2e60de0 fix(cost): split a provider-reported total between input and output by the cost map rate ratio (#49357)
b2647116ca chore(cost-map): add openai gpt-6-astra-law from the pricing page (#49444)
8ff495be96 fix(proxy): price OpenAI pass-through /v1/decisions calls like native Decisions (#49392)
be4f3d36ce feat(accounting): capture request usage through the durable meter (#49133)
Compare 36 commits »
scriptoriumadmin synced commits to refs/pull/49005/head at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
256e87789e Merge remote-tracking branch 'origin/main' into HEAD
c43a1addf2 refactor(router): extract PythonCore behind RouterCore and pass the router its config (#49119)
a5f980600c fix(a2a): answer proxy refusals on /a2a POST routes as JSON-RPC errors (#49476)
7143aeeb86 feat(passthrough): add Deepgram streaming text-to-speech WebSocket route (#49282)
3129e44848 ci: cut the weekly rc branch on Monday 3am PT instead of Friday (#49472)
Compare 49 commits »
scriptoriumadmin synced commits to refs/pull/48942/head at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
f418a83f49 fix(proxy): keep main's budget hold on generateContent routes
scriptoriumadmin synced commits to refs/pull/48946/head at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
4cbd830315 fix(guardrails): keep a pipeline step's guardrail list out of litellm_metadata, name Lakera v1 on a block, inject Lakera v2's client
8e26d9af5d Merge origin/main into litellm_fix_native_guardrails_on_extraction
61efbd071e fix(aiohttp): bill the usage a stream sent before its connection dropped, and deflake the CyberArk override test (#47838)
02a6a96dec feat(accounting): add bounded transactional usage persistence (#49190)
e6b4774ac6 fix(bedrock): restore the eu.anthropic.claude-haiku-4-5-20251001-v1:0 model map row (#49375)
Compare 14 commits »
scriptoriumadmin synced commits to refs/pull/49000/head at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
2642c1b06a fix(router): drop the oldest same-id least-busy attempt so running calls keep their start time
scriptoriumadmin synced commits to refs/pull/49000/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
2642c1b06a fix(router): drop the oldest same-id least-busy attempt so running calls keep their start time
60fc121ce8 test(a2a): replace the third-party agent e2e test with an integration test (#49541)
1b3da63703 test: move Sail, Weave and Presidio e2e tests to integration tests (#49533)
187f936cb3 fix(proxy)!: support reload with multiple workers (#45909)
Compare 162 commits »
scriptoriumadmin synced commits to refs/pull/49005/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
256e87789e Merge remote-tracking branch 'origin/main' into HEAD
c43a1addf2 refactor(router): extract PythonCore behind RouterCore and pass the router its config (#49119)
a5f980600c fix(a2a): answer proxy refusals on /a2a POST routes as JSON-RPC errors (#49476)
7143aeeb86 feat(passthrough): add Deepgram streaming text-to-speech WebSocket route (#49282)
Compare 48 commits »
scriptoriumadmin synced commits to refs/pull/49020/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
3129e44848 ci: cut the weekly rc branch on Monday 3am PT instead of Friday (#49472)
d610dee7d0 docs(pr-template): add a handwritten 1-3 sentence description section (#49475)
ced995df6e fix(logging): drop a first-token time left over from a previous attempt (#47541)
3155aacda4 fix(caching): stop counting an unrun server tool call mark, and follow #49192 and #35602 in their integration tests (#49468)
Compare 96 commits »
scriptoriumadmin synced commits to refs/pull/49021/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
3129e44848 ci: cut the weekly rc branch on Monday 3am PT instead of Friday (#49472)
d610dee7d0 docs(pr-template): add a handwritten 1-3 sentence description section (#49475)
ced995df6e fix(logging): drop a first-token time left over from a previous attempt (#47541)
3155aacda4 fix(caching): stop counting an unrun server tool call mark, and follow #49192 and #35602 in their integration tests (#49468)
Compare 96 commits »
scriptoriumadmin synced commits to refs/pull/49032/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:07 +00:00
a713ed6bab fix(gemini): add thought_signature_in_tool_call_id to keep tool call ids short for clients that cap them (#47692)
febd340120 fix(guardrails)!: keep the upstream error when a /v1/messages stream ends in one (#49088)
27fbcd9895 fix(anthropic): map pause_turn and return refusal explanations on chat and responses (#49192)
281bf4e48b ci(circleci): clone the public repo over HTTPS so main jobs stop failing at checkout (#49394)
Compare 96 commits »
scriptoriumadmin synced commits to refs/pull/48842/merge at scriptoriumadmin/litellm from mirror 2026-11-01 07:04:06 +00:00
60fc121ce8 test(a2a): replace the third-party agent e2e test with an integration test (#49541)
1b3da63703 test: move Sail, Weave and Presidio e2e tests to integration tests (#49533)
187f936cb3 fix(proxy)!: support reload with multiple workers (#45909)
a1c5829f6b fix(bedrock): raise GovCloud West Llama 3 max output to the model card 8K (#49543)
Compare 268 commits »