scriptoriumadmin
  • Joined on 2026-03-01
scriptoriumadmin synced commits to refs/pull/49176/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:32 +00:00
ff3983c1f6 fix(wandb): set DeepSeek-V3.1 and Llama-3.1-8B max_input_tokens from the CoreWeave Forge page (#49616)
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 35 commits »
scriptoriumadmin synced commits to refs/pull/49170/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:32 +00:00
408773bb69 test(spend): assert Anthropic envelope for Messages budget refusals (#49800)
3da6029977 test(bedrock): assert native structured output when the model rejects forced tools (#49799)
23515cf5b4 test(realtime): pin the relay failure exit with a client frame the provider rejects
fcef9cd0a3 feat(videos): serve custom_provider_map BaseVideoConfig providers on /v1/videos (#49785)
Compare 73 commits »
scriptoriumadmin synced commits to refs/pull/49170/head at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:32 +00:00
23515cf5b4 test(realtime): pin the relay failure exit with a client frame the provider rejects
cedad9ddf9 fix(realtime): count a relay that fails on its own as an abnormal end
fb6ad8db2f test(rate-limit): pin the split hold-release flags across nested MCP pre-calls
07cca4dbf0 fix(rate-limit): track the ITPM/OTPM hold's release apart from TPM so a nested pre-call keeps the request's project IO charge
2aceca2871 fix(rate-limit): take a token hold before awaiting so concurrent success and failure callbacks settle it once
Compare 5 commits »
scriptoriumadmin synced commits to refs/pull/49197/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:32 +00:00
acd084ab38 fix(mcp): answer bare OPTIONS on MCP routes without running LiteLLM auth (#49497)
81969e4afc refactor(ui): move admin and settings text onto i18n (#49546)
b84493d9a7 fix(proxy): log each endpoint's own call type on failed audio, moderation, image, and rerank spend rows (#49684)
105b7cba16 refactor(ui): move agents and knowledge text onto i18n (#49530)
Compare 200 commits »
scriptoriumadmin synced commits to refs/pull/49196/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:32 +00:00
01d6812f30 fix(model-catalog): declare the thinking_on_by_default key (#49720)
4e268c2c55 test(gemini): assert vertex lists on content chunks (#49734)
a2d27881e8 fix(bedrock): reserve DeepSeek V3.2 output capacity and date GovCloud West Llama 3 (#49732)
7fe6b1de34 fix(proxy)!: ignore stored general_settings keys no settings API manages (#48967)
Compare 91 commits »
scriptoriumadmin synced commits to refs/pull/49189/head at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:32 +00:00
de6371cc29 fix(redis-client): match redis-py command and failure behavior
4363955396 fix(auth): align cloud credential loading and expiry with the Python SDKs
044bd824c0 test(integration): align remaining release qualification checks (#49714)
05b76878b0 fix(cohere): map Cohere v2 finish_reason on non-streaming chat responses (#49529)
a4de720144 fix(count_tokens): count grok-4-1, grok-4.3 and DeepSeek V4.1 tool requests the way their hosts bill them (#49074)
Compare 23 commits »
scriptoriumadmin synced commits to refs/pull/49115/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:31 +00:00
6be98d8c75 fix(bedrock)!: give Invoke and Mantle Claude the model's own default max_tokens (#49692)
f8f12a270a test(integration): repair stale contract fixtures (#49700)
508a98586e fix(databricks): join plain text blocks for non-Claude json_object requests (#49385)
f91774e2f9 fix(fireworks): keep developer items in place for models that accept a later system message (#49399)
Compare 19 commits »
scriptoriumadmin synced commits to refs/pull/49110/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:31 +00:00
10b3deae0d build(deps): allow huggingface-hub 2.x (#49056)
cc142a8191 fix(cost): add deepgram/streaming/nova-3-medical and update stale Nova-3 streaming rates (#49733)
250af8f027 fix(spend)!: save streamed /v1/responses spend rows under the encrypted resp_ id the client received (#49493)
4ede2c012f feat(model_info): add supports_continuation capability flag for Gemini long decode continuation (#49500)
Compare 241 commits »
scriptoriumadmin synced commits to refs/pull/49109/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:31 +00:00
216041e243 feat(decisions): add Parallel AI web-grounded decisions (#49188)
08b951a814 fix(vllm): list hosted_vllm wildcard models when the server has no api key (#40066)
2a94bb3835 fix(spend): bill web search interception follow-ups once under the outer request (#49762)
be50cbe9d8 fix(responses)!: authorize and decode comparison response IDs (#49402)
Compare 63 commits »
scriptoriumadmin synced commits to refs/pull/49104/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:31 +00:00
10b3deae0d build(deps): allow huggingface-hub 2.x (#49056)
cc142a8191 fix(cost): add deepgram/streaming/nova-3-medical and update stale Nova-3 streaming rates (#49733)
250af8f027 fix(spend)!: save streamed /v1/responses spend rows under the encrypted resp_ id the client received (#49493)
4ede2c012f feat(model_info): add supports_continuation capability flag for Gemini long decode continuation (#49500)
Compare 34 commits »
scriptoriumadmin synced commits to refs/pull/49096/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:31 +00:00
10b3deae0d build(deps): allow huggingface-hub 2.x (#49056)
cc142a8191 fix(cost): add deepgram/streaming/nova-3-medical and update stale Nova-3 streaming rates (#49733)
250af8f027 fix(spend)!: save streamed /v1/responses spend rows under the encrypted resp_ id the client received (#49493)
4ede2c012f feat(model_info): add supports_continuation capability flag for Gemini long decode continuation (#49500)
Compare 153 commits »
scriptoriumadmin synced commits to refs/pull/49092/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:31 +00:00
0ea7e96a66 test(integration): repair agent reads and bulk deletion setup (#49719)
b4a80d175c ci: remove workflows whose secrets are dead (#49722)
c720b23238 feat(proxy)!: honor model_info.max_input_tokens in the Codex catalog for stock models (#49701)
b3c397e964 feat(hosted_vllm)!: keep vLLM's stop_reason and token_ids on chat choices and stream chunks (#48393)
Compare 86 commits »
scriptoriumadmin synced commits to refs/pull/49086/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:31 +00:00
10b3deae0d build(deps): allow huggingface-hub 2.x (#49056)
cc142a8191 fix(cost): add deepgram/streaming/nova-3-medical and update stale Nova-3 streaming rates (#49733)
250af8f027 fix(spend)!: save streamed /v1/responses spend rows under the encrypted resp_ id the client received (#49493)
4ede2c012f feat(model_info): add supports_continuation capability flag for Gemini long decode continuation (#49500)
Compare 34 commits »
scriptoriumadmin synced commits to refs/pull/49128/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:31 +00:00
6be98d8c75 fix(bedrock)!: give Invoke and Mantle Claude the model's own default max_tokens (#49692)
f8f12a270a test(integration): repair stale contract fixtures (#49700)
508a98586e fix(databricks): join plain text blocks for non-Claude json_object requests (#49385)
f91774e2f9 fix(fireworks): keep developer items in place for models that accept a later system message (#49399)
Compare 19 commits »
scriptoriumadmin synced commits to refs/pull/49130/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:31 +00:00
10b3deae0d build(deps): allow huggingface-hub 2.x (#49056)
cc142a8191 fix(cost): add deepgram/streaming/nova-3-medical and update stale Nova-3 streaming rates (#49733)
250af8f027 fix(spend)!: save streamed /v1/responses spend rows under the encrypted resp_ id the client received (#49493)
4ede2c012f feat(model_info): add supports_continuation capability flag for Gemini long decode continuation (#49500)
Compare 91 commits »
scriptoriumadmin synced commits to refs/pull/49132/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:31 +00:00
f106a09102 test(integration): isolate stranger batch authorization from the batch-cost poller (#49522)
1563776cd8 fix(prometheus): retain deployment attribution on proxy failures (#49717)
044bd824c0 test(integration): align remaining release qualification checks (#49714)
05b76878b0 fix(cohere): map Cohere v2 finish_reason on non-streaming chat responses (#49529)
Compare 109 commits »
scriptoriumadmin synced commits to refs/pull/49138/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:31 +00:00
6be98d8c75 fix(bedrock)!: give Invoke and Mantle Claude the model's own default max_tokens (#49692)
f8f12a270a test(integration): repair stale contract fixtures (#49700)
508a98586e fix(databricks): join plain text blocks for non-Claude json_object requests (#49385)
f91774e2f9 fix(fireworks): keep developer items in place for models that accept a later system message (#49399)
Compare 19 commits »
scriptoriumadmin synced commits to refs/pull/49145/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:31 +00:00
508a98586e fix(databricks): join plain text blocks for non-Claude json_object requests (#49385)
f91774e2f9 fix(fireworks): keep developer items in place for models that accept a later system message (#49399)
a2ab27373f fix(proxy): count the context_management edits /v1/messages applies on every bridged provider (#49234)
97cd5d0c6a fix(model-catalog): declare the tool and Responses token count framing keys (#49683)
Compare 16 commits »
scriptoriumadmin synced commits to refs/pull/49162/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:31 +00:00
10b3deae0d build(deps): allow huggingface-hub 2.x (#49056)
cc142a8191 fix(cost): add deepgram/streaming/nova-3-medical and update stale Nova-3 streaming rates (#49733)
250af8f027 fix(spend)!: save streamed /v1/responses spend rows under the encrypted resp_ id the client received (#49493)
4ede2c012f feat(model_info): add supports_continuation capability flag for Gemini long decode continuation (#49500)
Compare 34 commits »
scriptoriumadmin synced commits to refs/pull/49098/merge at scriptoriumadmin/litellm from mirror 2026-11-03 07:25:31 +00:00
10b3deae0d build(deps): allow huggingface-hub 2.x (#49056)
cc142a8191 fix(cost): add deepgram/streaming/nova-3-medical and update stale Nova-3 streaming rates (#49733)
250af8f027 fix(spend)!: save streamed /v1/responses spend rows under the encrypted resp_ id the client received (#49493)
4ede2c012f feat(model_info): add supports_continuation capability flag for Gemini long decode continuation (#49500)
Compare 34 commits »