mateo-berri
ba74b6a3b6
fix(ui): adapt the jev dashboard pieces to stable/1.102.x
...
(cherry picked from commit 4bfac88281
2026-09-23 00:41:28 +00:00
moe-berri
4bfd03e3de
feat(auto-router): add JEV classifier alongside LLM classifier
...
Backport of #41886 to stable/1.101.x.
Cherry-picked from a83773cfa5
2026-09-22 23:49:19 +00:00
moe-berri
02f386a82e
fix(proxy): enforce virtual key budgets for JEV test routing
...
Backport of #41879 to stable/1.101.x.
Cherry-picked from 1e161f516c
2026-09-22 23:15:11 +00:00
Yassin Kortam
8c30093c5a
feat(guardrails): add TypeSafe Jev relevance-based compaction guardrail
...
Backport of #41757 to stable/1.101.x.
Cherry-picked from 2edda5aec3
2026-09-22 23:12:35 +00:00
yuneng-jiang
eb0eca5451
fix(proxy): forward every method on the typesafe pass-through route
...
Backport of #41723 to stable/1.101.x.
Cherry-picked from 34718f0da6
2026-09-22 23:12:01 +00:00
Mateo Wang
2f63249816
feat(router): add TypeSafe Jev as a complexity router classifier
...
Backport of #41615 to stable/1.101.x.
Cherry-picked from cf42b607c3
2026-09-22 23:09:35 +00:00
Tin Chi Lo
3b97c28ab5
feat(auto-router): allow opted-in team members to manage their routers
...
Prerequisite for #41615 and #41879 on stable/1.101.x.
Cherry-picked from 109ca70f66
2026-09-22 23:09:31 +00:00
ryan-crabbe-berri
9250fd579c
fix(proxy): apply team model aliases on the JWT auth path
...
Prerequisite for #41615 on stable/1.101.x.
Cherry-picked from 7015bf37bb
2026-09-22 23:09:31 +00:00
mateo-berri
081f65ea44
feat(typesafe): add TypeSafe Jev passthrough with logging and cost tracking
...
Backport of #41607 to stable/1.101.x.
Cherry-picked from deb9d8aedd
2026-09-22 21:34:24 +00:00
yuneng-jiang
1fb7e5a9e9
Merge pull request #42332 from BerriAI/litellm_backport_1_101_x_bp_41870
...
chore(release): backport #41870 to stable/1.101.x
2026-09-21 14:47:16 -07:00
Yuneng Jiang
5b1a903976
chore(deps): bump soupsieve to 2.9.0
2026-09-21 14:18:31 -07:00
Yuneng Jiang
07a3aef0b2
chore(deps): bump anyio to 4.14.2
2026-09-21 14:18:30 -07:00
Mateo Wang
aec3026860
Merge pull request #41870 from BerriAI/litellm_bedrock_openai_gpt_min_max_tokens
...
fix(bedrock): clamp maxTokens to the 16-token minimum for OpenAI GPT and xAI Grok models on Converse
(cherry picked from commit a6e3a72ed8
2026-09-21 14:18:30 -07:00
Mateo Wang
8892c21b09
Merge pull request #41704 from BerriAI/litellm_bump_1_101_1
...
LiteLLM Rust / rustfmt, clippy, test (push) Has been cancelled
LiteLLM Rust / release wheel (push) Has been cancelled
chore(release): bump stable/1.101.x to 1.101.1
2026-09-17 17:21:25 -07:00
mateo-berri
53ee737406
chore(release): bump stable/1.101.x to 1.101.1
...
Sets the version project-releaser reads for the v1.101.1 stable cut, which carries #41698 (the wildcard license auto_router fix backported from #41684 ).
2026-09-17 17:00:23 -07:00
Mateo Wang
751fddecae
Merge pull request #41698 from BerriAI/litellm_cherrypick_1_101_x
...
fix(license): backport #41684 to stable/1.101.x so a wildcard license grants auto_router
2026-09-17 16:46:50 -07:00
mateo-berri
c9ed764c58
fix(license): let a wildcard allowed_features license grant the auto_router feature
...
Backport of #41684 to stable/1.101.x.
Cherry-picked from c2fbb11dca
2026-09-17 16:27:10 -07:00
yuneng-jiang
18243cd7af
Merge pull request #40886 from BerriAI/litellm_backport_redis_chaos_rc_1_101_0
...
LiteLLM Rust / rustfmt, clippy, test (push) Has been cancelled
LiteLLM Rust / release wheel (push) Has been cancelled
fix(redis): backport Redis chaos fixes and load gate to rc/1.101.0
2026-09-12 14:12:48 -07:00
Yuneng Jiang
abfb2cf329
test(load): backport Redis chaos qualification to rc/1.101.0 ( #40482 )
...
(cherry picked from commit 8e4f2abb40
2026-09-12 12:28:44 -07:00
Yuneng Jiang
c1a834baaf
fix(caching): backport Redis breaker recovery guards to rc/1.101.0 ( #40624 )
...
(cherry picked from commit ff4b558243
2026-09-12 12:28:18 -07:00
Yuneng Jiang
ca4a61304f
fix(redis): backport quiet breaker refusals to rc/1.101.0 ( #40620 )
...
(cherry picked from commit ef1a37795c
2026-09-12 12:27:57 -07:00
Yuneng Jiang
c324a2b3c5
perf(proxy): backport pipelined spend counters to rc/1.101.0 ( #40371 )
...
(cherry picked from commit 996ee5635a
2026-09-12 12:27:18 -07:00
yuneng-jiang
912a3f88f8
Merge pull request #40346 from BerriAI/litellm_backport_40335_rc_1_101_0
...
LiteLLM Rust / rustfmt, clippy, test (push) Has been cancelled
LiteLLM Rust / release wheel (push) Has been cancelled
fix(otel): backport auth spans and callback merge to rc/1.101.0 (#40335 )
2026-09-08 18:31:32 -07:00
Yuneng Jiang
bcddffcbdc
fix(otel): backport auth spans and callback merge to rc/1.101.0
...
Cherry-pick PR #40335 , restoring the Datadog auth span and last-wins callback credential merging.
(cherry picked from commit 43a1b2992a
2026-09-08 18:28:11 -07:00
yuneng-jiang
f0b6d66b84
Merge pull request #40325 from BerriAI/litellm_backport_37667_rc_1_101_0
...
feat(team): backport team admin callbacks to rc/1.101.0 (#37667 )
2026-09-08 16:42:17 -07:00
yuneng-jiang
f41fc719c1
Merge pull request #40324 from BerriAI/litellm_backport_40312_rc_1_101_0
...
chore(ui): backport dashboard dependencies to rc/1.101.0 (#40312 )
2026-09-08 16:42:09 -07:00
Yuneng Jiang
b6e73cccdb
feat(team): backport team admin callbacks to rc/1.101.0
...
Backport #37667 without conflict resolution or implementation changes
(cherry picked from commit 90731576e3
2026-09-08 16:35:28 -07:00
Yuneng Jiang
d8b9fcf399
chore(ui): backport dashboard dependencies to rc/1.101.0
...
Backport #40312 without conflict resolution or implementation changes
(cherry picked from commit 8280d7ca9d
2026-09-08 16:35:28 -07:00
yuneng-jiang
cbf0c76869
Merge pull request #40321 from BerriAI/litellm_backport_39654_rc_1_101_0
...
feat(otel): backport tenant trace destinations to rc/1.101.0 (#39654 )
2026-09-08 16:34:13 -07:00
Yuneng Jiang
09976043ac
feat(otel): backport tenant trace destinations to rc/1.101.0
...
Backport #39654 without conflict resolution or implementation changes
(cherry picked from commit 3165951e6d
2026-09-08 16:29:16 -07:00
yuneng-jiang
744156d297
Merge pull request #40316 from BerriAI/litellm_backport_40203_rc_1_101_0
...
feat: backport MongoDB sidecar to rc/1.101.0
2026-09-08 15:56:32 -07:00
yuneng-jiang
935bb5260f
feat: backport MongoDB sidecar to rc/1.101.0
...
(cherry picked from commit b8d573c5f9
2026-09-08 15:52:58 -07:00
ryan-crabbe-berri
eeb7732fc1
Merge pull request #39994 from BerriAI/litellm_hotfix_mongodb_extra_docker_images
...
Unit Tests / caching-local (push) Has been cancelled
Unit Tests / core-utils (push) Has been cancelled
Unit Tests / enterprise-routing (push) Has been cancelled
Unit Tests / integrations (push) Has been cancelled
Unit Tests / All Other Providers (push) Has been cancelled
Unit Tests / Vertex AI (push) Has been cancelled
Unit Tests / misc (push) Has been cancelled
Unit Tests / proxy-auth (push) Has been cancelled
Unit Tests / proxy-infra (push) Has been cancelled
Unit Tests / proxy-server (push) Has been cancelled
Terraform Modules / fmt, validate, test (gcp) (push) Has been cancelled
Terraform Provider / gofmt, vet, build, test (push) Has been cancelled
Unit Tests: Documentation Validation / documentation (push) Has been cancelled
GitHub Actions Security Analysis / zizmor (push) Has been cancelled
Unit Tests / enterprise-package (push) Has been cancelled
Unit Tests / proxy-endpoints (push) Has been cancelled
Unit Tests / proxy-extras (push) Has been cancelled
Unit Tests / responses-caching-types (push) Has been cancelled
Unit Tests: Proxy DB Operations / auth-checks (push) Has been cancelled
Unit Tests: Proxy DB Operations / budgets (push) Has been cancelled
Unit Tests: Proxy DB Operations / custom-logging (push) Has been cancelled
Unit Tests: Proxy DB Operations / db-and-spend (push) Has been cancelled
Unit Tests: Proxy DB Operations / endpoints-and-responses (push) Has been cancelled
Unit Tests: Proxy DB Operations / guardrails-hooks (push) Has been cancelled
Unit Tests: Proxy DB Operations / jwt-and-keys (push) Has been cancelled
Unit Tests: Proxy DB Operations / key-generation (push) Has been cancelled
Unit Tests: Proxy DB Operations / logging-misc (push) Has been cancelled
Unit Tests: Proxy DB Operations / proxy-runtime (push) Has been cancelled
Unit Tests: Proxy DB Operations / proxy-server-core (push) Has been cancelled
Unit Tests: Proxy DB Operations / proxy-utils (push) Has been cancelled
fix(docker): ship pymongo in the proxy images for the MongoDB vector store
2026-09-05 18:12:56 -07:00
Yuneng Jiang
16fd14f537
fix(docker): ship pymongo in the proxy images for the MongoDB vector store
...
The MongoDB Atlas vector store provider imports pymongo lazily from the
opt-in `mongodb` extra, but none of the shipped images installed that
extra. Any image-based deployment that configured a MongoDB vector store
failed at search time with "requires the 'pymongo' package", which the
user cannot fix without extending the image
Adds `--extra mongodb` to every uv sync in the root Dockerfile,
Dockerfile.database, Dockerfile.non_root, and the gateway component
image. The backend component does not serve /vector_stores so it is left
as is. The extra resolves from the existing uv.lock to pymongo 4.17.0
plus dnspython 2.8.0, no lock change needed
2026-09-05 18:03:24 -07:00
yuneng-jiang
e56b26c255
Merge pull request #39849 from BerriAI/litellm_internal_staging
...
chore(ci): promote internal staging to main
2026-09-05 16:41:13 -07:00
yuneng-jiang
bde58f9b41
Merge pull request #39959 from BerriAI/litellm_/release-ui-build-e256a3
...
chore: rebuild Admin UI bundle for the next release
2026-09-05 16:40:13 -07:00
Yuneng Jiang
0e118342ac
chore: update Next.js build artifacts (2026-09-05 23:25 UTC, node v24.19.0)
2026-09-05 16:25:57 -07:00
yuneng-jiang
b1e2f5bc0b
Merge pull request #39969 from BerriAI/revert-39121-litellm_lazy_sdk_import
...
revert: perf: lazy-load SDK symbols so import litellm stays under 60 MB RSS (#39121 )
2026-09-05 16:22:55 -07:00
devin-ai-integration[bot]
9a5564f00e
fix(ui): read Usage Total Requests tile from gateway request counts ( #39963 )
2026-09-05 16:20:56 -07:00
Yuneng Jiang
eee162b317
fix(lazy_imports): type import_map as Mapping to stay under the LIT001 budget
...
The revert restored the dict annotation that #39121 had loosened to Mapping,
and the LIT001 ceiling has been ratcheted down since, so the gate rejected the
one reintroduced hit. Annotation only, no behavior change.
2026-09-05 16:12:45 -07:00
ryan-crabbe-berri
587311d0e4
Merge pull request #39968 from BerriAI/litellm_remove_dead_model_hub_public_dialog
...
fix(ui): remove unreachable AI Hub dialog that put the session key in a URL
2026-09-05 16:10:15 -07:00
yuneng-jiang
45cf1a7ef1
Revert "perf: lazy-load SDK symbols so import litellm stays under 60 MB RSS (…"
...
This reverts commit c091dd4608
2026-09-05 16:07:09 -07:00
ryan-crabbe-berri
79a5614c18
fix(ui): remove unreachable AI Hub dialog that put the session key in a URL
...
The Public Model Hub dialog in ModelHubTable was never opened (its open setter had no callers), but its See Page button navigated to /model_hub_table?key=<session key>. Delete the dialog, its state, the handler and the unused router import so the path cannot be revived.
2026-09-05 16:04:50 -07:00
yuneng-jiang
1b25132863
Merge pull request #39953 from BerriAI/litellm_/litellm-e2e-flaky-test-2159ae
...
test(e2e): judge /v1/messages streaming on the clock, not on the provider's delta count
2026-09-05 16:04:45 -07:00
ryan-crabbe-berri
a0c9095cfc
Merge pull request #39960 from BerriAI/litellm_cache_hit_guardrail_spend
...
fix(proxy): keep guardrail cost in spend on cache hits
2026-09-05 15:53:34 -07:00
ryan-crabbe-berri
63156a7bd6
test(proxy): explain the proxy_server patches in the cache-hit regression test
...
The test-quality gate counts every patch of a litellm internal against a ceiling, and the three patches this test needs pushed it over. The callback imports increment_spend_counters, update_cache and proxy_logging_obj from proxy_server inside its own body, so there is no seam to inject fakes through; every other test in this file uses the same three patches for the same reason
Claude-Session: https://claude.ai/code/session_01EX13mWex6RaBo9PYnkAtFW
2026-09-05 15:44:37 -07:00
ryan-crabbe-berri
acddd21860
fix(proxy): keep guardrail cost in spend on cache hits
...
The proxy cost callback zeroed response_cost whenever cache_hit was true. That rule dates from Jan 2024 when it was the only place cache hits were priced. The logging layer has priced the LLM share at 0 on a cache hit since Aug 2024, and since guardrail cost joined the standard logging payload the proxy-side zeroing has thrown away a real provider charge: a pre_call guardrail runs before the cache is consulted, so a cached response still cost whatever the guardrail billed. Drop the redundant zeroing so the payload's response_cost, which is already LLM 0 + guardrail cost, reaches spend logs, daily tables and budgets untouched
Claude-Session: https://claude.ai/code/session_01EX13mWex6RaBo9PYnkAtFW
2026-09-05 15:32:02 -07:00
Mateo Wang
bf51dea36b
Merge pull request #39862 from BerriAI/litellm_lit_6992_cohere_parse
...
feat(ocr): add Cohere Parse support for cohere and azure_ai
2026-09-05 15:16:16 -07:00
yuneng-jiang
6a4fb2bbe8
Merge pull request #39938 from BerriAI/litellm_e2e_vertex_cache_first_call
...
test(e2e): prove Vertex context caching on the first cold call and on the spend row
2026-09-05 15:10:15 -07:00
Yuneng Jiang
cd976624d1
test(e2e): drop the explanatory sentence from the StreamingResponse docstring
2026-09-05 14:53:12 -07:00
