scriptoriumadmin

scriptoriumadmin synced commits to refs/pull/46882/merge at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:41 +00:00

91cfdb45ce Merge e6cfe62280 into 820725661e

820725661e fix(router): log each dropped /v1/messages attempt through its own logging copy and open the retry without its usage (#45194)

1fcbb0355e fix(pricing): update baseten reflection/Beam-1.0 prices from the models API (#47784)

045fb7ba84 feat(vertex-ai)!: support gemini 3.1 flash tts (#31915)

4ecc5fd13a feat(azure): add azure_ai/MAI-Code-1.1-Flash (#47779)

Compare 148 commits »

scriptoriumadmin synced commits to refs/pull/46883/merge at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:41 +00:00

d922e33a2f Merge 781f824974 into 8d6b8dab3d

8d6b8dab3d fix(cli): send pi's session id from lite pi so sessions group and stick (#47700)

8fe0a7358e ci(integration): keep the MCP SDK1 peer on mcp 1.x and autovacuum off the size cleanup test (#47651)

0e24bbb8a2 fix(responses): bill every round of a streamed MCP auto-execute on one spend row (#47445)

647ebd63e0 test(perf): emit per-location seconds, offending locations and the ten longest holds as event_loop_stalls breakdowns (#47757)

Compare 231 commits »

scriptoriumadmin synced commits to refs/pull/46888/merge at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:41 +00:00

1704474889 Merge be2489114f into 8d6b8dab3d

8d6b8dab3d fix(cli): send pi's session id from lite pi so sessions group and stick (#47700)

8fe0a7358e ci(integration): keep the MCP SDK1 peer on mcp 1.x and autovacuum off the size cleanup test (#47651)

0e24bbb8a2 fix(responses): bill every round of a streamed MCP auto-execute on one spend row (#47445)

647ebd63e0 test(perf): emit per-location seconds, offending locations and the ten longest holds as event_loop_stalls breakdowns (#47757)

Compare 227 commits »

scriptoriumadmin synced commits to refs/pull/46851/head at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:40 +00:00

e654f463e0 refactor(messages): log bridged /v1/messages streams when the client stream ends

5a50522bc5 refactor(responses): log bridged Responses streams on chat providers as Responses rows

8edddf5506 refactor(chat): log bridged chat streams on the Responses API as chat rows

e3a4ec7387 refactor(chat): keep responses-mode models on chat when the provider has no Responses API

2166698de6 refactor(responses): send bridged chat calls to the Responses executor

Compare 120 commits »

scriptoriumadmin synced commits to refs/pull/46852/head at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:40 +00:00

ddd506492e test(native): assert each cross-API bridge runs hooks, logging and the provider call once

e654f463e0 refactor(messages): log bridged /v1/messages streams when the client stream ends

5a50522bc5 refactor(responses): log bridged Responses streams on chat providers as Responses rows

8edddf5506 refactor(chat): log bridged chat streams on the Responses API as chat rows

e3a4ec7387 refactor(chat): keep responses-mode models on chat when the provider has no Responses API

Compare 121 commits »

scriptoriumadmin synced commits to refs/pull/46854/merge at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:40 +00:00

cb6169f551 Merge 4753536654 into 4b1928387f

4b1928387f test(bedrock): detect Vault reads on the event-loop thread (#47597)

208cac49f0 fix(vertex_ai)!: send integer enums to Gemini instead of dropping them (#44908)

ad8ca533ca feat(proxy): write one spend row per billed timed-out retry attempt (#47416)

2484f5dd6d test(integration): call the renamed metadata deployment helper in the Mantle count_tokens wire tests (#47592)

Compare 61 commits »

scriptoriumadmin synced commits to refs/pull/46857/head at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:40 +00:00

6d85aae12e fix(vertex_ai): copy request params per key before a cache retry so an uncopyable param no longer fails the call

d9b0e221f5 fix(vertex_ai): scope Vertex cache records by project and location, and drop a record a resume hop rejects

49837058bd fix(vertex_ai): let a timed-out waiter resolve on its own, reuse short-lived listed caches, and drop the record on any 4xx

9deeaa7ee7 fix(vertex_ai): keep the remembered cache on request-shape 400s, and stop a waiter at its own timeout

6a5fea1ad8 fix(vertex_ai): retry only errors that name the cache, and log the generateContent request after a rebuild

Compare 291 commits »

scriptoriumadmin synced commits to refs/pull/46859/merge at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:40 +00:00

7ff9ac5855 Merge d61c6c4d0d into 5ab0130abc

5ab0130abc test(integration): call _fallback_deployment in the count_tokens wire tests #47363 added (#47593)

9bbf70edb6 fix(completion): send prompt_cache_retention to OpenAI instead of silently dropping it (#47480)

8d2bf426f5 fix(cost): price base_model deployments past rule-only cost map entries (#47497)

dd22dc9132 test(perf): measure Admin UI journey latency in a browser on the 1TB SpendLogs restore (#47486)

Compare 54 commits »

scriptoriumadmin synced commits to refs/pull/46840/merge at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:40 +00:00

80a04127e1 Merge 4eb77d1ab2 into 047cb4a8e2

047cb4a8e2 test(integration): expect the Exa search envelope's provider_metadata after #47423 (#47572)

0ede0e10e7 fix(batch): preserve recoverable costs and prevent poll starvation (#47190)

ce63f48ed2 fix(decisions): log non-UTF-8 bytes as base64 so the request reaches validation and gets a 400 (#47564)

02199b670b fix(proxy): share the billing logger across callback registries (#47510)

Compare 52 commits »

scriptoriumadmin synced commits to refs/pull/46841/merge at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:40 +00:00

1c02751c28 Merge c1d02c6866 into 8d6b8dab3d

8d6b8dab3d fix(cli): send pi's session id from lite pi so sessions group and stick (#47700)

8fe0a7358e ci(integration): keep the MCP SDK1 peer on mcp 1.x and autovacuum off the size cleanup test (#47651)

0e24bbb8a2 fix(responses): bill every round of a streamed MCP auto-execute on one spend row (#47445)

647ebd63e0 test(perf): emit per-location seconds, offending locations and the ten longest holds as event_loop_stalls breakdowns (#47757)

Compare 227 commits »

scriptoriumadmin synced commits to refs/pull/46842/head at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:40 +00:00

8edddf5506 refactor(chat): log bridged chat streams on the Responses API as chat rows

e3a4ec7387 refactor(chat): keep responses-mode models on chat when the provider has no Responses API

2166698de6 refactor(responses): send bridged chat calls to the Responses executor

7aab20c882 refactor: call execute_chat from the Responses to chat bridge

5cbc378ddd refactor: plan forced-chat thinking on OpenAI straight to Responses

Compare 118 commits »

scriptoriumadmin synced commits to refs/pull/46845/merge at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:40 +00:00

4b807b5825 Merge 48c31361a5 into 7e8be61fa0

7e8be61fa0 test: fix flaky team delete, refusal sink, loop-blocking, Moyai tamper and model-cost leak tests (#47135)

e63aec029a feat(proxy): opt-in free-tier allowance for Google Search grounding spend (#47494)

b245cc0d26 feat(proxy): size-triggered spend log cleanup, oldest rows first (#47450)

4b1928387f test(bedrock): detect Vault reads on the event-loop thread (#47597)

Compare 69 commits »

scriptoriumadmin synced commits to refs/pull/46848/head at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:40 +00:00

5a50522bc5 refactor(responses): log bridged Responses streams on chat providers as Responses rows

8edddf5506 refactor(chat): log bridged chat streams on the Responses API as chat rows

e3a4ec7387 refactor(chat): keep responses-mode models on chat when the provider has no Responses API

2166698de6 refactor(responses): send bridged chat calls to the Responses executor

7aab20c882 refactor: call execute_chat from the Responses to chat bridge

Compare 119 commits »

scriptoriumadmin synced commits to refs/pull/46849/merge at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:40 +00:00

b02d0dd560 Merge fb373ada57 into 8d6b8dab3d

8d6b8dab3d fix(cli): send pi's session id from lite pi so sessions group and stick (#47700)

8fe0a7358e ci(integration): keep the MCP SDK1 peer on mcp 1.x and autovacuum off the size cleanup test (#47651)

0e24bbb8a2 fix(responses): bill every round of a streamed MCP auto-execute on one spend row (#47445)

647ebd63e0 test(perf): emit per-location seconds, offending locations and the ten longest holds as event_loop_stalls breakdowns (#47757)

Compare 227 commits »

scriptoriumadmin synced commits to refs/pull/46861/merge at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:40 +00:00

ed7b2570ef Merge 574d4b5670 into 8d6b8dab3d

8d6b8dab3d fix(cli): send pi's session id from lite pi so sessions group and stick (#47700)

8fe0a7358e ci(integration): keep the MCP SDK1 peer on mcp 1.x and autovacuum off the size cleanup test (#47651)

0e24bbb8a2 fix(responses): bill every round of a streamed MCP auto-execute on one spend row (#47445)

647ebd63e0 test(perf): emit per-location seconds, offending locations and the ten longest holds as event_loop_stalls breakdowns (#47757)

Compare 227 commits »

scriptoriumadmin synced commits to refs/pull/46779/merge at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:39 +00:00

2f92d444af Merge 7f0e9e9584 into 60e4afb6a6

60e4afb6a6 Merge e649ebcac4 into 0cf754b9b9

0cf754b9b9 Merge cd10e367ff into 5c341e7308

5c341e7308 test(perf): emit a top-contributors breakdown property per perf metric (#47632)

1380e7e921 feat(cost): add model_info.pricing_provider for cost-map lookups (#47620)

Compare 117 commits »

scriptoriumadmin synced commits to refs/pull/46799/merge at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:39 +00:00

501f900e88 Merge 8e509bf6dc into 9717b38050

9717b38050 bump: litellm-enterprise 0.1.78 -> 0.1.79, litellm-proxy-extras 0.4.110 -> 0.4.111, litellm 1.107.0 -> 1.108.0 (#47668)

5c836d6e13 feat(cost-map): add fireworks deepseek v4.1 flash 1017 us at the us-only premium (#47667)

8e3ce6d110 fix(vertex-ai): drop flex prices from gemini-nano-banana-2.1 (#47666)

6b1f4ed46f fix(logging): spilled over PTU requests without base_model log the served model's rates (#47655)

Compare 193 commits »

scriptoriumadmin synced commits to refs/pull/46800/merge at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:39 +00:00

5b3fe736d5 Merge 493c28a45c into 17db64ef78

17db64ef78 test(perf): record the row count and size of each table the Admin UI journeys query (#47683)

735ebc466c chore(deps-dev): bump vite from 8.3.0 to 8.3.4 in /vscode-extension (#47627)

6083cf9776 fix(mcp): list per-user auth MCP tools without a shared token (#47677)

49b5444ef9 fix(mcp): list per-user auth MCP tools without a shared token (#47677)

Compare 211 commits »

scriptoriumadmin synced commits to refs/pull/46809/head at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:39 +00:00

aff2f9dc43 refactor: extract execute_chat from completion()

7f0e9e9584 refactor: call execute_responses from the Messages to Responses bridge

e649ebcac4 refactor: extract execute_messages and execute_responses native executors

cd10e367ff refactor: decide the native API format once with plan_route

6e385f2f9f chore: remove description heading from PR template (#47631)

Compare 112 commits »

scriptoriumadmin synced commits to refs/pull/46809/merge at scriptoriumadmin/litellm from mirror 2026-10-21 05:17:39 +00:00

7d2c9f08b0 Merge aff2f9dc43 into 2f92d444af

2f92d444af Merge 7f0e9e9584 into 60e4afb6a6

60e4afb6a6 Merge e649ebcac4 into 0cf754b9b9

0cf754b9b9 Merge cd10e367ff into 5c341e7308

5c341e7308 test(perf): emit a top-contributors breakdown property per perf metric (#47632)

Compare 119 commits »