scriptoriumadmin
  • Joined on 2026-03-01
scriptoriumadmin synced commits to refs/pull/48940/merge at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:27 +00:00
a9b181489b fix(gemini): add the batch, flex and priority text output prices to gemini-2.5-flash-image (#49358)
458463a20d fix(search): forward Perplexity search filters (#49344)
969ded2b54 fix(guardrails): drop the response pipeline steps that never ran from x-litellm-applied-guardrails (#49264)
ef9385c712 refactor(proxy): inject gateway activity through a typed repository (#49116)
Compare 97 commits »
scriptoriumadmin synced commits to refs/pull/48958/head at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:27 +00:00
2a964a58d0 docs(skills): stores take domain inputs, callers keep per-kind failure outcomes
6a208ed271 feat(redis): cover every Python Redis use on the transport
77129e4f92 refactor(redis): move the response polling store into its own PR
2c69496b96 docs(skills): inline redis-access migration order
2db8c2ef94 docs(skills): make redis-access skill standalone
Compare 48 commits »
scriptoriumadmin synced commits to refs/pull/48948/merge at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:27 +00:00
faf9ee18e9 fix(cost): preserve zero prices across batch, speech and search (#49294)
7e5a653c01 chore(cost-map): add Deepgram Aura-1 and Aura-2 text-to-speech rates (#49280)
e8f572b39b feat(sso): create the team_id_jwt_field team on UI SSO login when team_id_upsert is set (#49256)
ec4f66bb94 fix(router): budget classifier input by provider limits (#49306)
Compare 129 commits »
scriptoriumadmin synced commits to refs/pull/48845/head at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:26 +00:00
6cd1f3d571 test(responses): preserve empty tool outputs through guardrails
6f6d5ff12a fix(responses): normalize empty function_call_output to an empty tool result
53ce41625f chore(databricks): retirement dates for DeepSeek V4 Flash (0731), V4 Pro (0813) and Inkling (#49124)
d25cde4792 chore(cost-map): add fireworks ultramarine 1 flash us and date deepseek v4 flash (#49146)
326f17594e fix(batches): return real batch token counts next to OpenAI's batch usage fields (#47740)
Compare 209 commits »
scriptoriumadmin synced commits to refs/pull/48895/merge at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:26 +00:00
a44e44b980 fix(cache_control): pick the chat-path dialect from the deployment's provider prefix (#45295)
87fec5467f fix(sso): keep CLI SSO and Claude Code gateway team lists in the user record's order (#49033)
02ef44c388 fix(streaming): convert upstream logprobs so streams with logprobs end with their finish chunk (#33549)
0070ce23b3 fix(router)!: apply content_policy_fallbacks when router settings update at runtime (#45428)
Compare 69 commits »
scriptoriumadmin synced commits to refs/pull/48882/merge at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:26 +00:00
4f7150f832 fix(ollama): forward assistant tool call ids in ollama_chat requests (#49278)
5898a42f5a fix(openrouter)!: list OpenRouter's live catalog for openrouter/* when check_provider_endpoint is on (#48820)
6ad4f9cfdf ci: cache Rust compilation with sccache (#49117)
b20acb4869 fix(proxy)!: persist sent model_info fields on POST /model/update (#46902)
Compare 60 commits »
scriptoriumadmin synced commits to refs/pull/48878/merge at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:26 +00:00
84617874b8 fix(ui): resolve auto router templates through ids a * catch-all lists over a partial wildcard (#49331)
8801a17f0d fix(proxy)!: bind admin-owned team keys to their team's MCP servers (#49161)
5950af8fb5 fix(responses): preserve text verbosity when bridging to chat completions (#49049)
61efbd071e fix(aiohttp): bill the usage a stream sent before its connection dropped, and deflake the CyberArk override test (#47838)
Compare 146 commits »
scriptoriumadmin synced commits to refs/pull/48867/merge at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:26 +00:00
fbd126e518 test: fix integration tests that fail on test order, races and intended changes (#49310)
632f71b850 fix(router): recover streamed context window errors before output (#49293)
f4d1f45d87 fix(proxy): raise live budget counters for failure spend rows (#47741)
8f968b6c8e feat(openrouter): record upstream provider in logging payload and spend logs (#49287)
Compare 161 commits »
scriptoriumadmin synced commits to refs/pull/48866/merge at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:26 +00:00
a42c1c138a fix(exceptions): classify SGLang context window overflow (#48837)
516c796e86 fix(rerank): preserve token usage in spend logs (#49308)
0382f134e7 fix(ui): add Meta Model API to the Add Model provider dropdown (#49314)
fbd126e518 test: fix integration tests that fail on test order, races and intended changes (#49310)
Compare 102 commits »
scriptoriumadmin synced commits to refs/pull/48863/merge at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:26 +00:00
84617874b8 fix(ui): resolve auto router templates through ids a * catch-all lists over a partial wildcard (#49331)
8801a17f0d fix(proxy)!: bind admin-owned team keys to their team's MCP servers (#49161)
5950af8fb5 fix(responses): preserve text verbosity when bridging to chat completions (#49049)
61efbd071e fix(aiohttp): bill the usage a stream sent before its connection dropped, and deflake the CyberArk override test (#47838)
Compare 196 commits »
scriptoriumadmin synced commits to refs/pull/48846/head at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:26 +00:00
b21b8ac5ab fix(decisions): count type values in a state as text, so a dropped image keeps a typed state's request
691b35c5f5 fix(decisions): count a state nested past the depth limit as no text, so its byte drops keep main's error
45d80099a8 Merge branch 'openrouter-decisions-images' of https://github.com/mtrefilek/litellm into openrouter-decisions-images
3d5a267434 fix(decisions): keep main's answer when a blank request's only download is a type the model refuses
96a80073eb fix(decisions): keep main's answer when a blank request loses its image to the declared type check, and count null as no text
Compare 113 commits »
scriptoriumadmin synced commits to refs/pull/48886/merge at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:26 +00:00
b7f17df47d fix(spend): read only the newest rows when paging sessions (LIT-8513)
837f36e872 fix(proxy): log router-refused image generation requests (#49349)
a9b181489b fix(gemini): add the batch, flex and priority text output prices to gemini-2.5-flash-image (#49358)
458463a20d fix(search): forward Perplexity search filters (#49344)
Compare 140 commits »
scriptoriumadmin synced commits to refs/pull/48886/head at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:26 +00:00
b7f17df47d fix(spend): read only the newest rows when paging sessions (LIT-8513)
scriptoriumadmin synced commits to refs/pull/48860/merge at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:26 +00:00
05bb13d996 fix(bedrock): consult the base model row when a regional CRIS row omits tier pricing
24b8889965 chore(cost-map): add openai gpt-6-rosalind from the pricing page and changelog (#49268)
a1c238ff4a refactor(rust): extract SecretValue and use it for Redis credentials (#49186)
d23af8bdd0 fix(cost): bill the margin once on rebuilt streams and streamed Responses through the chat bridge (#49254)
Compare 152 commits »
scriptoriumadmin synced commits to refs/pull/48860/head at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:26 +00:00
05bb13d996 fix(bedrock): consult the base model row when a regional CRIS row omits tier pricing
8928219d13 Merge remote-tracking branch 'origin/main' into litellm_bedrock_service_tier_drop_params
5d5e3685c1 fix(logging): warn when drop_params drops a requested reasoning level on the remaining paths (#48933)
df47ec0171 fix(spend): keep a NaN or infinite cost from wedging daily spend writes (#48829)
43c07a0f12 fix(router)!: check and count deployment itpm and otpm on every Responses WebSocket turn (#49069)
Compare 238 commits »
scriptoriumadmin synced commits to refs/pull/48846/merge at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:26 +00:00
a392a39627 fix(vertex_ai): bill Claude web search per request on deployments priced in model_info (#49160)
d1d6d55063 fix(hosted_vllm): accept the vLLM production-stack router's file and batch objects (#49068)
afec145edc feat(passthrough): let a WebSocket relay bill a session from its client text (#49281)
84617874b8 fix(ui): resolve auto router templates through ids a * catch-all lists over a partial wildcard (#49331)
Compare 134 commits »
scriptoriumadmin synced commits to refs/pull/48790/merge at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:25 +00:00
17d1f58336 fix(ui): preserve team model grants on unrelated settings saves (#49258)
372c096388 fix(proxy): evict a deleted key from every worker's auth cache (#47609)
ab3f6caddc fix(gemini): flag computer use support on the four models the Gemini computer use docs list (#49252)
9432bf6b8e perf(streaming): stop raising caught AttributeErrors inside pydantic on every streamed chunk (#49166)
Compare 72 commits »
scriptoriumadmin synced commits to refs/pull/48795/head at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:25 +00:00
9676865472 Merge branch 'main' into add-frequency-provider
bc24a606a0 feat(providers): Frequency gemma-4-26b-a4b-it max_output_tokens 8192
e6b4774ac6 fix(bedrock): restore the eu.anthropic.claude-haiku-4-5-20251001-v1:0 model map row (#49375)
b92222922a test(integration): match Databricks classifier budget telemetry (#49382)
98a1a6b209 fix(mcp): log SDK request rejections (#49184)
Compare 333 commits »
scriptoriumadmin synced commits to refs/pull/48795/merge at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:25 +00:00
9676865472 Merge branch 'main' into add-frequency-provider
bc24a606a0 feat(providers): Frequency gemma-4-26b-a4b-it max_output_tokens 8192
e6b4774ac6 fix(bedrock): restore the eu.anthropic.claude-haiku-4-5-20251001-v1:0 model map row (#49375)
b92222922a test(integration): match Databricks classifier budget telemetry (#49382)
Compare 334 commits »
scriptoriumadmin synced commits to refs/pull/48829/head at scriptoriumadmin/litellm from mirror 2026-10-31 07:12:25 +00:00
f5bb555720 Merge remote-tracking branch 'origin/main' into litellm_drop_non_finite_spend_cost
f32ac94ad1 fix(openai): keep the upstream error body on e.response for OpenAI and Azure errors (#49079)
4f65bb08c8 Merge remote-tracking branch 'origin/main' into litellm_drop_non_finite_spend_cost
ccc56d27ad fix(azure): add batch prices to the unversioned o3, o4-mini, o3-mini, gpt-4o and gpt-4o-mini rows (#49182)
b91fc92b6d fix(router): cap the default OTPM reservation and reconcile token counters after each call (#49054)
Compare 65 commits »