scriptoriumadmin
  • Joined on 2026-03-01
scriptoriumadmin synced commits to refs/pull/47776/merge at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:14 +00:00
3d7b94b34f fix(ollama): forward embedding credentials, headers and timeout (#48707)
4ce2602548 fix(huggingface): stop forcing truncate=false on TEI rerank and pass truncate params through (#48706)
323d3956d6 feat(rust): one typed decline from core up to each host (#47772)
7b60fd061a refactor(rust): adopt one error discipline across litellm-rust with source chains and Report (#48042)
Compare 90 commits »
scriptoriumadmin synced commits to refs/pull/47772/head at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:14 +00:00
89890398e2 merge: integrate upstream bridge boundary regression fix
e88597d785 test(rust): preserve bridge serialization boundary in body regression
9fea2cb01d merge: integrate full upstream bodies with bounded diagnostics
3a1a2017cc fix(rust): reconcile restacked base and body regression
d1b975e40c fix(http): normalize provider URLs before building outbound requests
Compare 295 commits »
scriptoriumadmin synced commits to refs/pull/47764/merge at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:14 +00:00
9bb31caa28 fix(guardrails): release an unfinished choice's held text and omit usage the upstream never sent
b7da9a2755 fix(credentials): keep the stored secret when PATCH echoes its masked read-back (#48950)
a66dc50efb test(json_fragment_accumulator): assert peak allocation instead of a drain timing ratio (#43370)
cbf4e40d38 fix(proxy)!: stop labelling management and health rejections as auth failures (#48739)
Compare 187 commits »
scriptoriumadmin synced commits to refs/pull/47764/head at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:14 +00:00
9bb31caa28 fix(guardrails): release an unfinished choice's held text and omit usage the upstream never sent
f07a549e13 fix(guardrails): replace blocked text completion streams like chat, keep Gray Swan's vendor payload, scan text streams in the content filter
1f814963f7 Merge branch 'main' into litellm_completions_stream_guardrail_scan
b96c1f30a6 feat(ocr)!: support outputContentFormat=markdown for Azure Document Intelligence (#48926)
a2ef18b257 test(unit): keep unix socket paths in unit tests under the macOS AF_UNIX limit (#45040)
Compare 409 commits »
scriptoriumadmin synced commits to refs/pull/47762/head at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:14 +00:00
09b49357cb fix: end text completion n=2 streams cleanly, keep Azure image gateway query params, tighten SDK 3 test fixtures
b913a0fdbf Merge remote-tracking branch 'origin/main' into litellm_openai_sdk3_docker
924ac590b1 fix(openai): end a dropped stream only after every requested choice finished, keep Azure deployment URL query params, and narrow scope
db5b940143 fix(vertex_ai): send Vertex Claude labels as X-Vertex-AI-Labels header instead of the request body (#48753)
cfc651dcd5 fix(http): honor transport settings for Rust WebSockets (#47968)
Compare 314 commits »
scriptoriumadmin synced commits to refs/pull/47759/head at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:13 +00:00
81b3107894 fix(vector_stores): refuse a caller-chosen opensearch_index_name on search and rag query
1a78d7ba6a fix(vector_stores): sign opensearch for the endpoint's region, refuse list, add opensearch_index_name, honor drop_params
6010a8050b test(vector_stores): inject a failing opensearch credential source instead of running a shell
f1166f5ed8 fix(vector_stores): return only scalar opensearch attributes and keep credential source output out of 401s
db240fbbc6 fix(vector_stores): keep opensearch retrieve, update, and delete off the index and refuse half-set AWS keys and per-account endpoints
Compare 345 commits »
scriptoriumadmin synced commits to refs/pull/47759/merge at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:13 +00:00
81b3107894 fix(vector_stores): refuse a caller-chosen opensearch_index_name on search and rag query
cbf4e40d38 fix(proxy)!: stop labelling management and health rejections as auth failures (#48739)
a11e76fe8f fix(google_genai): read one-object and null spellings of repeated fields in countTokens and generateContent (#48771)
764a7a6536 fix(context_management)!: count the system prompt toward the clear_tool_uses input_tokens trigger (#47857)
Compare 187 commits »
scriptoriumadmin synced commits to refs/pull/47761/head at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:13 +00:00
812f0054a4 fix(reasoning): suggest reasoning_effort_levels and base_model only where the route reads them
f5f5715ea7 fix(reasoning): warn when any reasoning-off request is dropped and give each route only the advice it honors
9eaba58c73 fix: stay silent when reasoning is turned off on /v1/responses and warn on the Bedrock Responses fallback drop
d32087cd8e fix: skip the drop warning when a non-reasoning model is asked for no reasoning, and honor declared levels on Bedrock Responses
00d74f7196 Merge commit 'c7c895db456eda349a7be9661b4b21f578f9b589' into litellm_warn_dropped_reasoning_effort
Compare 268 commits »
scriptoriumadmin synced commits to refs/pull/47760/head at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:13 +00:00
52c658e3ea Merge remote-tracking branch 'origin/main' into litellm_bedrock_strict_tools_retry
e32d8a4d03 fix(responses): reject or drop Responses params the chat completions bridge cannot translate (#48655)
df9091deb6 fix(responses): skip failed spend-log turns when rebuilding a session from previous_response_id (#48647)
c7c895db45 fix(proxy)!: paginate customer lists and usage filters (#48700)
7acdf9f00d fix(anthropic): count search results and reject unsupported token-count blocks (#48699)
Compare 260 commits »
scriptoriumadmin synced commits to refs/pull/47754/head at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:13 +00:00
ca6505fd58 Merge remote-tracking branch 'origin/main' into litellm_semantic_cache_gate_hook
c9a5777739 fix(azure): share one Entra token request per worker and audit #47348 (#48769)
a45c2023ce fix(audio): return 400 for unsupported transcription providers (#47971)
a0c9a5993b fix(batches): stop unbillable batches from blocking file deletes and new-batch polling (#48752)
78d6c1aa82 fix(router): pass lifecycle events through on background OpenAI responses streams that cannot fall back (#48817)
Compare 318 commits »
scriptoriumadmin synced commits to refs/pull/47753/head at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:13 +00:00
8c8bcff43f Merge remote-tracking branch 'origin/main' into litellm_managed_files_hooks_wheel_install
dfd9ad4103 fix(responses)!: bill partial usage on mid-stream failures and report bridged /v1/messages failures to the proxy (#47537)
a604c89059 fix(ui): bundle the Aria avatar so it loads in every deployment (#48890)
aa769fbc69 fix(streaming): surface upstream abort as server errors (LIT-10159) (#48328)
1f68cc5649 refactor(rust-bridge): require the native extension (#48852)
Compare 355 commits »
scriptoriumadmin synced commits to refs/pull/47751/merge at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:13 +00:00
0454ef8efb Merge remote-tracking branch 'origin/main' into litellm_bedrock_embed_region_prefix
be649f66a6 feat(azure_ai): reach Foundry agents through the project Responses API (#47717)
b96c1f30a6 feat(ocr)!: support outputContentFormat=markdown for Azure Document Intelligence (#48926)
a2ef18b257 test(unit): keep unix socket paths in unit tests under the macOS AF_UNIX limit (#45040)
Compare 179 commits »
scriptoriumadmin synced commits to refs/pull/47751/head at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:13 +00:00
0454ef8efb Merge remote-tracking branch 'origin/main' into litellm_bedrock_embed_region_prefix
56e60d3b7d feat(router): daily_reset_time picks when deployment rpd and tpd daily limits start over (#48957)
f46adf80e4 fix(proxy): let typesafe/* allowlists cover /typesafe Jev calls and list Jev in the model picker (#47697)
3a350cacc9 feat!: add StreamLake provider metadata (#41006)
ce2813a669 fix(bedrock): keep chat's region rules for embeddings and leave ARN pricing as on main
Compare 410 commits »
scriptoriumadmin synced commits to refs/pull/47750/head at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:13 +00:00
e0ef7dfa1e Merge remote-tracking branch 'origin/main' into litellm_vertex_passthrough_single_auth
13386b9550 fix(proxy): size a Gemini request's budget hold from its prompt and maxOutputTokens
66ddbaab86 fix(ci): spend logs guard requires the approval to cover the head's spend-log lines (#48904)
4516a251c9 feat(accounting): add accounting contracts, PricingSnapshot and ledger schema (LIT-10193) (#48862)
cf7c35a117 fix(anthropic): map model_context_window_exceeded stop reason to length (#48874)
Compare 369 commits »
scriptoriumadmin synced commits to refs/pull/47756/merge at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:13 +00:00
dbaae521f9 fix(responses): drop the null created_by OpenAI rejects when output items are replayed (#48783)
b1b5d16f0b docs(adrs): add ADRs for the Python SDK and proxy, starting with provider usage extras (#36668)
be649f66a6 feat(azure_ai): reach Foundry agents through the project Responses API (#47717)
b96c1f30a6 feat(ocr)!: support outputContentFormat=markdown for Azure Document Intelligence (#48926)
Compare 175 commits »
scriptoriumadmin synced commits to refs/pull/47749/head at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:12 +00:00
db8bcc999a Merge remote-tracking branch 'origin/main' into litellm_health_scope_like_auth
cbf4e40d38 fix(proxy)!: stop labelling management and health rejections as auth failures (#48739)
a11e76fe8f fix(google_genai): read one-object and null spellings of repeated fields in countTokens and generateContent (#48771)
764a7a6536 fix(context_management)!: count the system prompt toward the clear_tool_uses input_tokens trigger (#47857)
de42b67252 fix(proxy)!: enforce an end user's per-model budget from their linked budget (#48815)
Compare 422 commits »
scriptoriumadmin synced commits to refs/pull/47741/merge at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:12 +00:00
83ae1bd63e fix(responses): drop params compact rejects and send Azure compaction to the v1 route (#48876)
4fc2bba1f2 fix(stream): the last usage frame carries litellm's bill in usage.cost (#48955)
019a9de603 fix(logging): stop dumping the request body at every layer under --detailed_debug (#48973)
57f05e56f6 fix(bedrock): omit system prompt from audio transcription and report its token usage (#41565)
Compare 174 commits »
scriptoriumadmin synced commits to refs/pull/47740/merge at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:12 +00:00
cfc651dcd5 fix(http): honor transport settings for Rust WebSockets (#47968)
6c0585b8f1 build(rust): upgrade reqwest to 0.13 and move MCP out of litellm-http (#48039)
38b4d31447 test(scim): audit #47544 (#48791)
356069d2ed test(mcp): audit #47422 (#48821)
Compare 71 commits »
scriptoriumadmin synced commits to refs/pull/47738/head at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:12 +00:00
7846534c7b chore: merge main
e4ddb5d281 chore: merge main and follow the mcp_server package move
ddb8a14ee9 feat(skills): add a litellm-setup agent skill (#47731)
41c4c3788e test(routing): audit #46601 (#48910)
567383c503 feat(proxy)!: model family and access group entries in model_max_budget (#47722)
Compare 375 commits »
scriptoriumadmin synced commits to refs/pull/47731/head at scriptoriumadmin/litellm from mirror 2026-10-29 06:36:12 +00:00
dce2cc6380 docs(skills): price Azure deployments with base_model and put uv on PATH in the start block
970603032d test(skills): find non-end-user skills in-process instead of through git
335e252e13 fix(skills): hide every non-end-user skill, use Azure deployment names, keep the restart block out of the user's shell
e189d5683b docs(skills): point litellm --setup users to the master key in litellm_config.yaml
014137bd12 fix(skills): give the litellm --setup path a restart command that keeps its config
Compare 355 commits »