scriptoriumadmin
  • Joined on 2026-03-01
scriptoriumadmin synced commits to refs/pull/45039/head at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:43 +00:00
7e0663a6ad fix(proxy): apply the worker log config before the proxy app import, as earlier releases did
39c9748c92 fix(gateway): ship litellm_cli in the gateway image
ea88f23d93 fix(proxy): load the worker app before its log config and without GC pauses
0565affb6b Merge remote-tracking branch 'origin/main' into litellm_worker_healthcheck_default
71babd931b fix(llms): move later developer messages for models declared system-first (#48897)
Compare 88 commits »
scriptoriumadmin synced commits to refs/pull/45030/merge at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:43 +00:00
c98641fd5a feat(dashscope)!: derive protocol bases from one domain (#47812)
06cdee91da fix(guardrails): scan and rewrite streamed /v1/completions under post_call guardrails (#49035)
961fabbb76 feat(secret_managers): resolve oidc/google api_key references per request (#47703)
6cdb9959df feat(router): opt in to non-chat deployment filtering (#46257)
Compare 230 commits »
scriptoriumadmin synced commits to refs/pull/44838/merge at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:42 +00:00
71babd931b fix(llms): move later developer messages for models declared system-first (#48897)
d037d86a94 fix(token_counter)!: keep a loaded Hugging Face tokenizer when a slower download fails (#48916)
c87a25a357 feat(proxy)!: general_settings.claude_code_marketplace_auth_required puts the Claude Code marketplace behind a LiteLLM key (#48980)
e8f0735d8a feat(proxy): let a Claude Code gateway user choose the session team (#49001)
Compare 90 commits »
scriptoriumadmin synced commits to refs/pull/44863/head at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:42 +00:00
e8ab6a0d1f fix(openai_like): route Hostinger natives per model from providers.json, refuse non-text embeddings headers unless drop_params
c1d0b8f16c fix(openai_like): send Hostinger's 1h cache TTL as the 5-minute default so cache writes bill at its charge
6e3be1f44d Merge remote-tracking branch 'origin/main' into litellm_pr44863_head
77064d6c70 feat(vector_stores): add OpenSearch k-NN search as a vector store provider (#47759)
be6d082bdc fix(anthropic)!: send /v1/messages PDF documents to OpenAI-shaped providers as file parts (#47166)
Compare 172 commits »
scriptoriumadmin synced commits to refs/pull/44858/merge at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:42 +00:00
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)
594f862bf3 fix(streaming): rebuild each choice of an n>1 chat stream on its own (#49002)
Compare 324 commits »
scriptoriumadmin synced commits to refs/pull/44854/merge at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:42 +00:00
83d6779158 fix(guardrails)!: make lakera_v2 on_flagged block reject PII-only detections, add mask (#48944)
a0a8ca97d9 fix(cost): finalize synchronous response cost once for header, spend log and rollups (#48993)
a1bd5feebe feat(router): infer custom classifier tiers from model mappings (#48987)
5dcc3a1bbf docs: link README to Python SDK and autorouter guides (#48986)
Compare 293 commits »
scriptoriumadmin synced commits to refs/pull/44863/merge at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:42 +00:00
7aa0393cb6 feat(providers): add CLF AI Gateway as an OpenAI-compatible provider with ten models (#39324)
fd594ab0cf fix: enforce LiteAdmin MCP route permissions (#48080)
e0778bc9dd refactor(types): replace Any with proven types in 21 files (#48921)
19e538ab11 fix(cost-map): set together_ai Qwen3.7-Max back to the published price (#49125)
Compare 213 commits »
scriptoriumadmin synced commits to refs/pull/44867/merge at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:42 +00:00
3f01341f0a fix(health)!: openai realtime health check reports healthy with a missing or invalid api key (#41233)
904c092ffe feat(providers): add BigModel support and model pricing (#48563)
9a68573e9b fix(responses): keep each Codex tool once when its tool list repeats across turns (#49006)
3c1eb73a06 fix(proxy)!: list every priced Vertex family under a vertex_ai/* grant (#41959)
Compare 163 commits »
scriptoriumadmin synced commits to refs/pull/44723/head at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:41 +00:00
d9f37b67b3 test(github_copilot): type the authenticator test fixture and the tests this branch adds
scriptoriumadmin synced commits to refs/pull/44784/head at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:41 +00:00
fcc88b195b fix(anthropic): drop an unmappable context_management list under drop_params
f44af58a7a Merge remote-tracking branch 'origin/main' into litellm_count_tokens_compaction_block
4550fc7ac2 fix(anthropic): forward count context_management only with edits, map Vertex list-form betas, and return provider context_management
77064d6c70 feat(vector_stores): add OpenSearch k-NN search as a vector store provider (#47759)
be6d082bdc fix(anthropic)!: send /v1/messages PDF documents to OpenAI-shaped providers as file parts (#47166)
Compare 69 commits »
scriptoriumadmin synced commits to refs/pull/44822/head at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:41 +00:00
446d33df70 fix(soniox): let a server VAD turn answer only the oldest pending commit
cf4dc70aea Merge remote-tracking branch 'origin/main' into litellm_soniox_realtime_stt
b93d26bc85 fix(responses): drop a malformed prompt_cache_breakpoint under drop_params on the direct /v1/responses path (#45330)
310c96162b fix(anthropic): drop the non-spec custom object from /v1/messages tool definitions (#48893)
42d2909210 test(integration): scope the signoz, otel conversation id, and series cap proxy fixtures to their module (#48965)
Compare 61 commits »
scriptoriumadmin synced commits to refs/pull/44784/merge at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:41 +00:00
68b5a737a1 fix(anthropic): retry /v1/messages without strict when Claude rejects strict tools under drop_params (#49102)
08b539e6a8 fix(proxy)!: bill streamed /v1/completions without stream_options like chat (#45366)
e93e01ca4b fix(ci): build base native extension for OpenAPI lint (#49136)
fd8b89fbec fix(mcp)!: restore upstream-token admission for legacy delegate servers (#48939)
Compare 85 commits »
scriptoriumadmin synced commits to refs/pull/44576/merge at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:41 +00:00
93c5cebf9d fix(guardrails): answer ThirdLaw streamed Responses and Messages blocks with their block status
72232ea88a Merge remote-tracking branch 'upstream/main' into feature/thirdlaw-guardrail
ccc56d27ad fix(azure): add batch prices to the unversioned o3, o4-mini, o3-mini, gpt-4o and gpt-4o-mini rows (#49182)
be649272d4 test(guardrails): pin that a ThirdLaw streamed block reaches the client as its status
Compare 331 commits »
scriptoriumadmin synced commits to refs/pull/44712/merge at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:41 +00:00
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)
342ce6ae4a feat(zai): support response_format and max_completion_tokens, add glm-5v-turbo pricing (#48335)
733cfeba93 fix(cost): bill a provider reported usage.cost on every path unless the deployment is priced (#48953)
Compare 134 commits »
scriptoriumadmin synced commits to refs/pull/44747/head at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:41 +00:00
e247c5221f Merge remote-tracking branch 'origin/main' into litellm_vertex_compact_2026_09_04_beta
c98641fd5a feat(dashscope)!: derive protocol bases from one domain (#47812)
06cdee91da fix(guardrails): scan and rewrite streamed /v1/completions under post_call guardrails (#49035)
961fabbb76 feat(secret_managers): resolve oidc/google api_key references per request (#47703)
6cdb9959df feat(router): opt in to non-chat deployment filtering (#46257)
Compare 151 commits »
scriptoriumadmin synced commits to refs/pull/44772/head at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:41 +00:00
d39df0d4ac fix(usage): keep a stored subscription fee listed after its account's deployments are removed
32fe58ad00 fix(usage): count a subscription fee only for accounts behind a current deployment
572cb1faa3 fix(usage): merge main and count ChatGPT Responses calls without a default login
c213234783 feat(vertex_ai)!: use HTTP/2 httpx client for search_api vector store (#40631)
ead221995e refactor(router): move router callers onto the entry points and invert the proxy imports (#48847)
Compare 119 commits »
scriptoriumadmin synced commits to refs/pull/44772/merge at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:41 +00:00
dbcacc7ab7 fix(health): probe only models a partial wildcard serves (#49060)
f32ac94ad1 fix(openai): keep the upstream error body on e.response for OpenAI and Azure errors (#49079)
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 152 commits »
scriptoriumadmin synced commits to refs/pull/44747/merge at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:41 +00:00
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)
342ce6ae4a feat(zai): support response_format and max_completion_tokens, add glm-5v-turbo pricing (#48335)
733cfeba93 fix(cost): bill a provider reported usage.cost on every path unless the deployment is priced (#48953)
Compare 204 commits »
scriptoriumadmin synced commits to refs/pull/44783/head at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:41 +00:00
b80fd9bc93 fix(vertex_ai): use the stricter cache minimum and send a Gemini base's 404 uncached
scriptoriumadmin synced commits to refs/pull/44368/merge at scriptoriumadmin/litellm from mirror 2026-10-30 06:47:40 +00:00
d8f162ea22 fix(guardrails): allow MCP call modes on the generic guardrail API (#48060)
1a2b443a32 ci: run the extensions integration group with 48 workers (#49134)
8d441ebc15 test(otel): give the Langfuse destination tests the otel callback's stdout exporter (#49129)
4ce724e3df fix(caching): key cached async clients by a per-loop token instead of id(loop) (#48877)
Compare 303 commits »