BerriAI/litellm v1.104.0-dev.1 on GitHub

Verify Docker Image Signature

All LiteLLM Docker images are signed with cosign. Every release is signed with the same key introduced in commit 0112e53.

Verify using the pinned commit hash (recommended):

A commit hash is cryptographically immutable, so this is the strongest way to ensure you are using the original signing key:

cosign verify \
  --key https://raw.githubusercontent.com/BerriAI/litellm/0112e53046018d726492c814b3644b7d376029d0/cosign.pub \
  ghcr.io/berriai/litellm:v1.104.0-dev.1

Verify using the release tag (convenience):

Tags are protected in this repository and resolve to the same key. This option is easier to read but relies on tag protection rules:

cosign verify \
  --key https://raw.githubusercontent.com/BerriAI/litellm/v1.104.0-dev.1/cosign.pub \
  ghcr.io/berriai/litellm:v1.104.0-dev.1

Expected output:

The following checks were performed on each of these signatures:
  - The cosign claims were validated
  - The signatures were verified against the specified public key

What's Changed

build(deps): re-suppress GHSA-h7x2-h6g9-p789 in osv-scan, mlflow still has no fixed release by @devin-ai-integration[bot] in #41036
feat(auth): add breached password detection to password policy by @ojensen-berri in #39321
feat(auth): add self-service password change and plug plaintext password leaks by @ojensen-berri in #39562
feat(auth): force password reset for breached or admin-set passwords by @ojensen-berri in #40107
perf(proxy): serialize /model/info listing once with orjson by @devin-ai-integration[bot] in #41061
fix(utils): stop wrapper_async submitting the sync success handler twice by @devin-ai-integration[bot] in #41058
test(cost_map): stop pinning supports_reasoning absent on the openrouter o1 entry by @devin-ai-integration[bot] in #41734
fix(model_prices): add cache-read pricing to Mistral chat models missing it by @devin-ai-integration[bot] in #41736
refactor(rust): isolate legacy callback contract by @devin-ai-integration[bot] in #41752
test(cost_map): drop tests that pin provider-owned catalog values by @devin-ai-integration[bot] in #41763
test(e2e/ui): wait for the filtered budget list before clicking a row action by @yuneng-berri in #41756
chore(prices): sync OpenRouter prices: 443 models, 191 new, 4 deprecated by @berriai-litellm-provider-info-sync[bot] in #41727
fix(batches): bill Bedrock Titan embedding batch lines from inputTextTokenCount by @devin-ai-integration[bot] in #41767
fix(proxy): persist only the keys a caller changed in save_config by @yuneng-berri in #41748
chore(prices): sync OpenRouter prices: 2 models, 1 deprecated by @berriai-litellm-provider-info-sync[bot] in #41770
fix(proxy): run prompt injection heuristics off the event loop by @devin-ai-integration[bot] in #41541
fix(proxy): evict jwt key mapping cache on user, team, org, and bulk key deletion by @devin-ai-integration[bot] in #41707
refactor(ui): remove unused HelpLink and HelpIcon components by @devin-ai-integration[bot] in #41644
chore(ui): remove orphaned ROLE_STYLES and RoleStyle from pretty messages view by @devin-ai-integration[bot] in #41645
chore(ui): remove never-rendered GuardrailConfig mock component and its test by @devin-ai-integration[bot] in #41646
chore(ui): remove unused NewBadge component and its test by @devin-ai-integration[bot] in #41647
chore(ui): remove dead compareUI ModelSelector and its test by @devin-ai-integration[bot] in #41649
refactor(ui): remove unused createCredentialFromModel helper and CredentialValues interface by @devin-ai-integration[bot] in #41650
refactor(ui): drop unused cost-tracking barrel re-exports and response types by @devin-ai-integration[bot] in #41651
refactor(ui): drop unused rolesAllowedToSeeUsage, viewOnlyRoles and isViewOnlyRole exports by @devin-ai-integration[bot] in #41653
chore(ui): remove unused access-groups type interfaces by @devin-ai-integration[bot] in #41655
refactor(ui): remove dead networking exports and orphaned Claude Code marketplace helpers by @devin-ai-integration[bot] in #41656
refactor(ui): remove dead useKeyList hook from key_list.tsx by @devin-ai-integration[bot] in #41657
chore(ui): remove orphaned useDeleteProject hook and its test by @devin-ai-integration[bot] in #41658
refactor(interactions): remove expired use_legacy_interactions_schema shim by @devin-ai-integration[bot] in #41663
refactor(rust): align crates with Python package layering by @devin-ai-integration[bot] in #41829
chore(prices): sync OpenRouter prices: 15 models, 6 deprecated by @berriai-litellm-provider-info-sync[bot] in #41772
refactor(proxy): make the config file win over the database by @yuneng-berri in #41779
fix(ui): let admins change a model's team from the model edit page by @devin-ai-integration[bot] in #40700
fix(ui): show per-second pricing for video models instead of $0.00 token costs by @devin-ai-integration[bot] in #39308
fix(cost): bill cache-read tokens at the input rate when the map has no cache-read rate by @devin-ai-integration[bot] in #41832
fix(responses): honor nested additional_drop_params paths by @devin-ai-integration[bot] in #40730
feat(router): reject with 429 when a deployment's max_parallel_requests slots are all in use by @devin-ai-integration[bot] in #41555
perf: defer fastapi and tiktoken BPE imports out of import litellm by @devin-ai-integration[bot] in #41585
feat(proxy): add Amazon Transcribe pass-through with completion-time job pricing by @devin-ai-integration[bot] in #41515
feat(proxy): per-key default budget for dynamically created customers by @devin-ai-integration[bot] in #41636
feat(mcp): show live gateway sessions by AI client and user by @devin-ai-integration[bot] in #41692
ci: replace the title-similarity duplicate bot with a Codex semantic check by @ryan-crabbe-berri in #40935
feat(proxy): temporary budget increase for team members by @devin-ai-integration[bot] in #41620
feat(vault): add separate login and secret namespaces for HashiCorp Vault by @devin-ai-integration[bot] in #41539
feat(vertex_ai): stream GCS batch output files from /v1/files/{id}/content by @devin-ai-integration[bot] in #41506
feat(a2a): reach Microsoft Foundry agents with Entra auth and versioned card discovery by @devin-ai-integration[bot] in #41511
fix(azure): strip litellm format field from file and image content parts by @devin-ai-integration[bot] in #41275
ci: classify new issues into domain, provider, kind, priority and lift labels by @ryan-crabbe-berri in #41695
fix(router): validate routing_groups at save time and keep invalid DB groups from blocking SSO load by @devin-ai-integration[bot] in #41351
fix(bedrock): send aws_session_tags on every STS call via one typed auth struct by @ryan-crabbe-berri in #40500
fix(anthropic): register thinking-binding-controls-2026-08-01 in beta headers config by @clonylu in #41203
feat(guardrails): add TypeSafe Jev relevance-based compaction guardrail by @devin-ai-integration[bot] in #41757
feat(keys): let team service account keys use key management endpoints for their own team by @devin-ai-integration[bot] in #40807
test(docs): read only the first column of the router_settings reference table by @devin-ai-integration[bot] in #41860
fix(scim): accept entitlements and roles entries without a value on SCIM user PUT by @devin-ai-integration[bot] in #41830
perf(proxy): split aggregated usage query into key-free rollups and bounded top-N keys by @devin-ai-integration[bot] in #41293
chore(deps): bump anyio from 4.13.0 to 4.14.2 by @dependabot[bot] in #41846
feat(proxy): limit repeated failed Admin UI sign-in attempts by @devin-ai-integration[bot] in #40982
fix(proxy): make SettingsStore.clear() terminate when the config file owns a key by @devin-ai-integration[bot] in #41862
feat(vscode): add LiteLLM language model provider extension by @devin-ai-integration[bot] in #41865
feat(mcp): let proxy admins force-close live MCP sessions and revoke stored user credentials by @devin-ai-integration[bot] in #41725
feat(models): add qwen3.8 flash rows, fix Cohere embed v3 context, Bedrock Mantle and OpenRouter pricing by @devin-ai-integration[bot] in #41754
fix(azure): keep api-version query after vector store search path by @devin-ai-integration[bot] in #41384
fix(router): honor stream_timeout on the SDK-native passthrough route (/v1/messages, /converse) by @devin-ai-integration[bot] in #41875
fix(proxy): track team member spend when the member has no budget by @devin-ai-integration[bot] in #41349
feat(proxy): add Azure AI Speech pass-through route by @devin-ai-integration[bot] in #41557
fix(responses): stop managed Responses WebSocket from leaking litellm_params into provider request body by @devin-ai-integration[bot] in #33101
feat(rust): port exception_type to litellm-core-utils by @devin-ai-integration[bot] in #41873
fix(ui): show internal user email in logs table and log detail drawer by @devin-ai-integration[bot] in #40737
fix(router): report null cost for unpriced deployments instead of 0 by @devin-ai-integration[bot] in #40878
feat(azure_ai): support FLUX.2 flex images by @emerzon in #39424
fix(mistral): accept reasoning_effort on all models and drop client_metadata for Codex compatibility by @devin-ai-integration[bot] in #41062
feat(passthrough): deepgram streaming /v1/listen WebSocket passthrough with duration-based cost tracking by @devin-ai-integration[bot] in #41554
test(ocr): declarative provider x auth x input matrix, fix Rust 401 mapping by @devin-ai-integration[bot] in #41884
fix(proxy): classify Azure Speech short audio behind a prefixed api base by @devin-ai-integration[bot] in #41882
feat(proxy): add LiteLLM_DailyGlobalSpend key-free rollup for the usage dashboard by @devin-ai-integration[bot] in #41324
fix(proxy): keep requested model guardrails and key disable_fallbacks on rate-limit fallback by @devin-ai-integration[bot] in #41783
fix(bedrock): carry s3_endpoint_url and s3_region_name into file content downloads by @devin-ai-integration[bot] in #41138
ci(duplicate-check): let Codex reach GitHub from its sandbox by @ryan-crabbe-berri in #41883
fix(proxy): resolve model_group_alias to its target for /v1/models metadata by @devin-ai-integration[bot] in #41483
fix(proxy): reset sibling tpm/rpm counters when the shared rate limit window rolls over by @devin-ai-integration[bot] in #41838
feat(ui): link MCP Servers page to the user's connected MCP servers by @devin-ai-integration[bot] in #41888
fix(proxy): name the blocking guardrail in x-litellm-applied-guardrails by @devin-ai-integration[bot] in #41583
fix(responses): merge deployment litellm_params into native websocket response.create frames by @devin-ai-integration[bot] in #41881
feat: honor eager_input_streaming on Bedrock and Anthropic Claude tools by @devin-ai-integration[bot] in #41871
fix(otel): keep caller traceparent and tracestate on pass-through relays by @devin-ai-integration[bot] in #41786
ci: remove the dead Agent Shin triage workflows and scripts by @devin-ai-integration[bot] in #41894
fix: set vertex gemma-4-26b-a4b-it-maas context window to 262144 by @devin-ai-integration[bot] in #41887
refactor(rust): formalize legacy callback contract by @devin-ai-integration[bot] in #41885
fix(gemini): preserve candidates with finishReason and no content (#40477) by @devin-ai-integration[bot] in #41892
fix(team): apply team_member_budget updates to members still on the team default by @devin-ai-integration[bot] in #41347
fix(enterprise): resolve openai_moderations model at call time and default to omni-moderation-latest by @devin-ai-integration[bot] in #41895
fix(proxy): track project spend and enforce project budgets additively by @devin-ai-integration[bot] in #41354
fix(proxy): refuse config-owned keys on POST /config/update by @devin-ai-integration[bot] in #41868
test(integration): cover off-peak pricing on a live proxy by @devin-ai-integration[bot] in #41901
fix(proxy): requeue daily spend rows when the commit fails without the Redis buffer by @devin-ai-integration[bot] in #41878
fix(responses): emit typed streaming failure events by @zoroyihan7 in #40243
fix(bedrock): clamp maxTokens to the 16-token minimum for OpenAI GPT and xAI Grok models on Converse by @devin-ai-integration[bot] in #41870
chore(model_info): backfill reseller Gemini entries from provider catalogs and prune retired ids by @devin-ai-integration[bot] in #41902
feat(mcp): allowlist MCP client applications at the gateway by @devin-ai-integration[bot] in #41667
fix(schema): classify off_peak_pricing as a structured object in the model prices schema generator by @devin-ai-integration[bot] in #41847
fix(rate_limiter): render the 429 reset time in UTC as labelled by @devin-ai-integration[bot] in #41911
ci(issues): comment which release carries the fix when a pull request closes an issue by @ryan-crabbe-berri in #41910
feat(proxy): let team admins manage projects via team_admin_editable_team_fields by @devin-ai-integration[bot] in #41916
fix(bedrock): sign batch retrieve and cancel with deployment credentials when AWS_BEARER_TOKEN_BEDROCK is set by @devin-ai-integration[bot] in #41904
fix(websearch_interception): surface a failed search as a web_search_tool_result_error block and end the turn by @devin-ai-integration[bot] in #41905
fix(cost_calc): default fireworks cached input to the documented 50% discount when the map has no cache-read rate by @devin-ai-integration[bot] in #41917
fix(timing): anchor response duration and overhead at proxy receive time by @devin-ai-integration[bot] in #41891
feat(xai): add speech-to-text (Grok Voice Transcribe) via /v1/audio/transcriptions by @devin-ai-integration[bot] in #41914
feat(rust): add litellm-http client pool and inject it into the OCR route by @devin-ai-integration[bot] in #41897
feat(vertex_ai): stream Chirp speech-to-text over /v1/realtime by @devin-ai-integration[bot] in #41721
chore: consolidate CLAUDE.md into AGENTS.md by @devin-ai-integration[bot] in #41921
feat(proxy): serve the Claude Code gateway protocol under /claude_code_gateway by @devin-ai-integration[bot] in #34267
fix(responses): restore encrypted_content and apply affinity on the native WebSocket relay by @devin-ai-integration[bot] in #41893
feat(proxy): add RFC 8693 token exchange for IdP JWTs on the gateway token endpoint by @devin-ai-integration[bot] in #41485
fix(websearch): forward the deployment api_base to agentic follow-up calls on /v1/messages by @devin-ai-integration[bot] in #41918
fix(proxy): parse role_permissions where it is read by @yuneng-berri in #41924
fix(responses): announce message item before text events in the chat completions bridge by @devin-ai-integration[bot] in #41564
fix(proxy): unpin cost-map pricing copied into model_info and report pricing overrides by @devin-ai-integration[bot] in #41843
feat(websearch): let the model emit objective + multi-query search shapes by @adssoccer1 in #40399
fix(exceptions): keep internal_server_error as the public type of an upstream 500 by @devin-ai-integration[bot] in #41930
fix(proxy): dispatch llm_api_check moderation through during_call_hook by @devin-ai-integration[bot] in #41685
fix(policy_engine): deliver guardrail text rewrites on multi-choice, unfinished, and envelope-less streams by @devin-ai-integration[bot] in #41933
fix(anthropic): keep cache_control for Gemini targets on /v1/messages and normalize Anthropic ttl units by @devin-ai-integration[bot] in #41938
fix(guardrails): stop the Javelin api_version default leaking into Azure Content Safety by @devin-ai-integration[bot] in #41941
fix(alerting): clarify budget threshold messages by @gaurav-pandey-zocdoc in #39102
fix(cost): carry image and video input tokens through the Responses usage bridge (internal copy of #36887) by @devin-ai-integration[bot] in #41237
ci(unit): fail a hung test in 120s with a traceback instead of idling the shard to its step timeout by @devin-ai-integration[bot] in #41948
refactor(types): replace Any with proven types in 6 files by @devin-ai-integration[bot] in #41947
fix(proxy): register transcribe as a known provider for model grants by @yuneng-berri in #41926
feat(batches): support Mistral files/batches and per-page OCR batch cost tracking (internal copy of #40484) by @devin-ai-integration[bot] in #41934
fix(utils): reject an untranslatable tool_choice with a 400 instead of a 500 by @devin-ai-integration[bot] in #41234
fix(proxy): keep request metadata out of the cost tracking failure alert by @devin-ai-integration[bot] in #41950
fix(masker): memoize shared nodes and fail closed past the depth cap by @devin-ai-integration[bot] in #41952
fix(responses): drop tool_search and local_shell in the chat completions bridge by @devin-ai-integration[bot] in #41953
fix(rag): resolve registry stores on /v1/rag/ingest and reject providers without ingestion by @devin-ai-integration[bot] in #41940
fix(azure): send the resolved Entra ID token on image generation requests by @abhirup7 in #40147
fix(proxy): keep the raw client model out of spend logs for rejections outside the router by @devin-ai-integration[bot] in #41943
fix(cost): resolve dated openai/azure snapshots to their undated cost map entry by @devin-ai-integration[bot] in #41423
fix(proxy): /key/bulk_update writes only the fields each item carries by @devin-ai-integration[bot] in #41949
fix(cost): bill DeepSeek V4.1 Flash and V4 Pro at off-peak rates outside peak hours by @devin-ai-integration[bot] in #41960
test(cost): point dated snapshot tests at a date the cost map cannot carry by @devin-ai-integration[bot] in #41971
fix(proxy): refuse runtime writes to config-owned settings by @yuneng-berri in #41931
refactor(rust): centralize layered settings resolution by @devin-ai-integration[bot] in #41969
test(utils): isolate dated model fallback from pricing additions by @joshua-berri in #41975
test(response_metadata): anchor detailed-timing test on a fixed instant instead of wall clock by @devin-ai-integration[bot] in #41946
test(integration): literal request/response cost tracking suite on a test-owned cost map by @devin-ai-integration[bot] in #41328
chore(prices): sync Azure prices: 5 models, 5 deprecated by @berriai-litellm-provider-info-sync[bot] in #41966
feat(ocr): add Rust-only Textract and sign provider requests after host hooks by @devin-ai-integration[bot] in #41977
refactor(rust): use typed pyo3 APIs instead of getattr/import strings by @devin-ai-integration[bot] in #41981
fix(router): enforce model tpm limits against shared redis usage across replicas by @devin-ai-integration[bot] in #41915
feat(proxy): say when a stored setting is ignored because the config file owns it by @yuneng-berri in #41985
fix(otel v2): summarize embedding vectors as Langfuse observation output by @devin-ai-integration[bot] in #41982
fix(caching): scope automatic breakpoints to supported Claude transports by @tin-berri in #41920
fix(proxy): return 400 instead of 500 for /v1/responses without input by @devin-ai-integration[bot] in #41939
fix(llmguard): accept proxy async call types by @yucheng-berri in #41787
fix(model_prices): drop anthropic deprecation floors and correct azure gpt-4.1-nano retirement date by @devin-ai-integration[bot] in #41964
fix(ui): show user attribution in Top Virtual Keys usage tables by @devin-ai-integration[bot] in #40729
chore(prices): sync OpenRouter prices: 172 models, 2 new by @berriai-litellm-provider-info-sync[bot] in #41833
fix(ocr): set DeepSeek OCR sampling defaults by @devin-ai-integration[bot] in #41992
ci: remove auto-merge-price-sync workflow, the Devin sync automation merges price PRs by @devin-ai-integration[bot] in #41993
refactor(mcp): upgrade SDK2 while preserving legacy gateway behavior by @joshua-berri in #41718
revert(guardrails): drop the scoped request conversation and tools from post-call scans (#41220) by @devin-ai-integration[bot] in #41986
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #41996
test(unified_google_tests): use the Vertex global endpoint and retry 429s with backoff by @devin-ai-integration[bot] in #41995
fix(rust): refuse native routes in processes forked after the runtime started by @devin-ai-integration[bot] in #41987
fix(terraform): unlink the registry docs entries that 404 on click by @shivamrawat1 in #42003
test(integration): native responses and messages cost cases by @devin-ai-integration[bot] in #42010
fix(otel v2): map Responses API output onto the Langfuse generation output by @devin-ai-integration[bot] in #41991
test(e2e): stop the config suite locking itself out of the shared proxy by @yuneng-berri in #42008
chore: bump litellm-enterprise 0.1.68 -> 0.1.69, litellm-proxy-extras 0.4.99 -> 0.4.100 by @yuneng-berri in #42025
fix(proxy): close the config-ownership gaps QA found in the settings store by @yuneng-berri in #42009
fix(proxy): block project requests when max_budget is 0 by @ryan-crabbe-berri in #41997
fix(auth): inherit org alias, budget and rate limits for JWT and team-linked keys by @devin-ai-integration[bot] in #41681
feat(batches): run hosted_vllm batches inside LiteLLM by @devin-ai-integration[bot] in #41942
fix(proxy): forward stream response attributes through the hook boundary and merge logged applied_guardrails by @devin-ai-integration[bot] in #42027
feat(ui): configure web search interception from the Admin UI by @yuneng-berri in #42007
fix(azure): drop tool_choice when the request has no tools (internal copy of #41781) by @devin-ai-integration[bot] in #42031
fix(azure): tool choice without tools causes 400 by @dsteeley in #41781
test(mcp): verify scoped execution and OAuth credential isolation by @joshua-berri in #41731
test(e2e): cover MCP OAuth SSO and cold restart persistence by @devin-ai-integration[bot] in #41909
fix(proxy): estimate auto-router baseline costs from durable cache history by @tin-berri in #41177
fix(auto-router): show heuristic v2 score estimates in routing details by @tin-berri in #42001
feat(router): add maintained Fuse model and harness presets by @tin-berri in #41617
refactor(rust): rename legacy callback adapter crate by @devin-ai-integration[bot] in #42037
Revert "test(mcp): verify scoped execution and OAuth credential isolation" by @mateo-berri in #42039
revert(e2e): remove MCP OAuth SSO and cold restart tests by @mateo-berri in #42040
test(mcp): cover SDK redirect compatibility by @joshua-berri in #42033
feat(ui): report whether the serving proxy has applied web search interception by @yuneng-berri in #42042
fix(test): unbreak the integration-cost and proxy_e2e_anthropic_messages CircleCI jobs on main by @devin-ai-integration[bot] in #42048
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42006
feat(otel v2): opt-in llm_only span scope for Langfuse destinations and the operator Langfuse exporter by @devin-ai-integration[bot] in #41740
test(e2e): point the Nova Sonic realtime test at nova-2-sonic by @yuneng-berri in #42053
fix(team): emit audit events for member_delete and role changes and carry the final roster on team create by @devin-ai-integration[bot] in #41840
chore(prices): sync OpenRouter prices: 5 models by @berriai-litellm-provider-info-sync[bot] in #42058
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42063
chore: rebuild Admin UI bundle from main (build kXnLzJ6ylsRPmgSkCkCKM) by @yuneng-berri in #42054
fix(proxy): stop re-sending un-resendable spend batches from the Redis buffer by @devin-ai-integration[bot] in #41994
fix(router): skip cooldown for background response cost poll 404s by @devin-ai-integration[bot] in #42046
fix(router): stamp model_group when retrieving a batch, so batch tokens are attributable (internal copy of #38499) by @devin-ai-integration[bot] in #42062
test(mcp): restore scoped execution and credential isolation regressions by @joshua-berri in #42050
feat(proxy)!: refuse to start with an unset, empty, or publicly known master key by @ryan-crabbe-berri in #42019
docs: stop advertising sk-1234 as the master key in shipped configs and examples by @ryan-crabbe-berri in #42011
test(logging): add autorouter estimate keys to the GCS pub/sub spend-log golden by @devin-ai-integration[bot] in #42061
fix(router): add NotFoundErrorRetries so a retry policy can pin 404 retries by @devin-ai-integration[bot] in #42045
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42077
fix(google_genai): forward response schema and tool parameters through the generateContent adapter by @devin-ai-integration[bot] in #42067
test(e2e): restore MCP OAuth happy-path coverage (LIT-3467) by @joshua-berri in #42051
fix(mcp): tools/call no longer 404s on a worker that has not served tools/list by @devin-ai-integration[bot] in #42072
chore(prices): sync OpenRouter prices: 7 models, 6 deprecated by @berriai-litellm-provider-info-sync[bot] in #42082
fix(proxy): wait for the spend-log table before creating startup views by @devin-ai-integration[bot] in #41974
fix(guardrails): scan each choice's tool-call arguments apart on n>1 streams and log why a rewrite was discarded by @devin-ai-integration[bot] in #40986
chore(prices): sync OpenRouter prices: 1 model by @berriai-litellm-provider-info-sync[bot] in #42089
chore(prices): sync OpenRouter prices: 1 model by @berriai-litellm-provider-info-sync[bot] in #42092
chore: remove the dead telemetry flag from the SDK, proxy CLI and configs by @devin-ai-integration[bot] in #42071
docs(tests): define the tier contract for unit, integration and e2e by @devin-ai-integration[bot] in #42099
fix(proxy): coordinate v2 migration startup and qualify container recovery by @yuneng-berri in #40932
ci(tests): wire tests/unit into CircleCI and keep draining GHA shards green by @devin-ai-integration[bot] in #42103
test(unit): block external sockets at import time and add a socket policy regression test by @devin-ai-integration[bot] in #42113
fix(ci): excuse retired test-quality rules in the budget ratchet by @devin-ai-integration[bot] in #42116
test: migrate wave 1 phase 2 legacy unit tests to tests/unit by @devin-ai-integration[bot] in #42108
test(llms): migrate bedrock, baseten and base_llm batch tests to tests/unit by @devin-ai-integration[bot] in #42110
test(llms): migrate phase 6 provider unit tests to tests/unit by @devin-ai-integration[bot] in #42107
test: migrate nvidia, oci, ocr, oobabooga and openai legacy tests to tests/unit by @devin-ai-integration[bot] in #42115
test: migrate phase 9 legacy llm provider tests to tests/unit by @devin-ai-integration[bot] in #42117
test(llms): migrate phase 7 provider unit tests to tests/unit by @devin-ai-integration[bot] in #42114
test: migrate wave 1 phase 8 legacy llm tests to tests/unit by @devin-ai-integration[bot] in #42112
test(llms): migrate phase 5 provider unit tests to tests/unit by @devin-ai-integration[bot] in #42109
refactor(types): replace Any with proven types in 30 files by @devin-ai-integration[bot] in #42127
test(unit): make every tests/unit directory a package so pytest collection is unique by @devin-ai-integration[bot] in #42135
test(unit): migrate wave 1 phase 3 anthropic, apiserpent, azure and azure_ai legacy tests by @devin-ai-integration[bot] in #42118
test: migrate openai, openai_like and openrouter legacy tests to tests/unit by @devin-ai-integration[bot] in #42128
test: migrate phase 14 wave 2 provider tests to tests/unit by @devin-ai-integration[bot] in #42132
test: migrate phase 16 legacy tests to tests/unit by @devin-ai-integration[bot] in #42131
test: migrate wave 1 phase 1 legacy tests to tests/unit by @devin-ai-integration[bot] in #42123
test: migrate phase 12 legacy llm provider tests to tests/unit by @devin-ai-integration[bot] in #42137
test: migrate phase 15 legacy tests to tests/unit by @devin-ai-integration[bot] in #42136
fix(proxy): enforce virtual key budgets for JEV test routing by @devin-ai-integration[bot] in #41879
feat(fal_ai): add Seedance 2.5 / 2.0 video generation via fal queue API by @devin-ai-integration[bot] in #41980
chore(prices): sync OpenRouter prices: 11 models by @berriai-litellm-provider-info-sync[bot] in #42096
test: migrate legacy provider tests to tests/unit (wave 2, phase 13) by @devin-ai-integration[bot] in #42145
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42154
chore(prices): sync OpenRouter prices: 5 models by @berriai-litellm-provider-info-sync[bot] in #42155
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42157
fix(auth): reject deactivated JWT users and refresh cached status by @joshua-berri in #42064
fix(mcp): explain missing public client dependencies by @joshua-berri in #42148
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42162
chore(prices): sync OpenRouter prices: 3 models by @berriai-litellm-provider-info-sync[bot] in #42163
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42164
chore(prices): sync OpenRouter prices: 1 model by @berriai-litellm-provider-info-sync[bot] in #42166
chore(prices): sync OpenRouter prices: 3 models by @berriai-litellm-provider-info-sync[bot] in #42168
refactor(rust): split token counter backends by @devin-ai-integration[bot] in #42165
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42169
chore(prices): sync OpenRouter prices: 3 models by @berriai-litellm-provider-info-sync[bot] in #42175
chore(prices): sync OpenRouter prices: 4 models by @berriai-litellm-provider-info-sync[bot] in #42178
chore(prices): sync OpenRouter prices: 3 models by @berriai-litellm-provider-info-sync[bot] in #42179
chore(prices): sync OpenRouter prices: 1 model by @berriai-litellm-provider-info-sync[bot] in #42182
feat(rust): add typed secret managers and shared auth adapters by @devin-ai-integration[bot] in #42173
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42184
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42187
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42192
fix(responses): forward safety_identifier through the chat completion bridge by @devin-ai-integration[bot] in #42193
refactor(types): replace Any with proven types in 32 files by @devin-ai-integration[bot] in #42220
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42227
fix(helm): render a fixed replicaCount on componentized deployments when HPA is disabled by @devin-ai-integration[bot] in #42207
fix(proxy): return 422 instead of 429 for BudgetExceededError by @devin-ai-integration[bot] in #42097
chore(prices): sync OpenRouter prices: 1 model by @berriai-litellm-provider-info-sync[bot] in #42234
chore(prices): sync OpenRouter prices: 1 model by @berriai-litellm-provider-info-sync[bot] in #42240
chore(prices): sync OpenRouter prices: 4 models by @berriai-litellm-provider-info-sync[bot] in #42243
fix(a2a): send message/stream for Bedrock AgentCore streaming requests by @devin-ai-integration[bot] in #42239
fix(registry): add MAI-Image-2.5-Pro pricing, fix Fireworks/Together entries, absorb verified open registry PRs, add Groq deprecation and Bedrock regional Qwen3 Next pricing by @devin-ai-integration[bot] in #34941
fix(anthropic): forward safeguards and anthropic-beta unchanged on native /v1/messages by @devin-ai-integration[bot] in #42152
chore(prices): sync OpenRouter prices: 4 models by @berriai-litellm-provider-info-sync[bot] in #42246
chore(prices): sync OpenRouter prices: 3 models by @berriai-litellm-provider-info-sync[bot] in #42251
chore(prices): sync Azure prices: 1 model, 1 deprecated by @berriai-litellm-provider-info-sync[bot] in #42249
chore(prices): sync AWS Bedrock prices: 25 models [enrichment failed: AWS Bedrock, 66 held] by @berriai-litellm-provider-info-sync[bot] in #42250
test(a2a): migrate a2a_protocol legacy tests to tests/unit (wave 3 phase 17) by @devin-ai-integration[bot] in #42160
feat(proxy): default to the v2 migration resolver by @yuneng-berri in #42105
feat(auto-router): configure heuristic v2 success threshold by @tin-berri in #42252
chore(prices): sync Together AI prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42258
chore(prices): sync OpenRouter prices: 6 models, 1 new by @berriai-litellm-provider-info-sync[bot] in #42253
fix(key_generate): use the user's own budget as the ceiling for UI session personal keys by @devin-ai-integration[bot] in #41588
feat(team): show whether a member follows the team default budget and allow resetting to it by @devin-ai-integration[bot] in #41906
chore(prices): sync AWS Bedrock prices: 13 models, 1 new [1 with gaps, enrichment failed: AWS Bedrock, 38 held] by @berriai-litellm-provider-info-sync[bot] in #42254
chore(prices): sync OpenRouter prices: 7 models by @berriai-litellm-provider-info-sync[bot] in #42261
feat(proxy): add GET /utils/model_info to look up cost map info for unregistered models by @devin-ai-integration[bot] in #42121
chore(prices): sync OpenRouter prices: 4 models by @berriai-litellm-provider-info-sync[bot] in #42265
feat(xai): add grok-4.7 to the cost map by @devin-ai-integration[bot] in #42264
ci(e2e): fix the stage-mirror batch reds and keep a redacted pytest log by @devin-ai-integration[bot] in #42143
chore(prices): sync AWS Bedrock prices: 3 models [enrichment failed: AWS Bedrock, 32 held] by @berriai-litellm-provider-info-sync[bot] in #42269
chore(prices): sync OpenRouter prices: 6 models by @berriai-litellm-provider-info-sync[bot] in #42270
feat(fal_ai): add gpt-image-2.5 flare/sunburst, flux/dev and image edits by @devin-ai-integration[bot] in #42095
feat(bedrock): add us.moonshotai.kimi-k3 pricing and fill the global Kimi K3 entry by @devin-ai-integration[bot] in #42271
test(google): boot the unified Google proxy fixture with a real master key by @yuneng-berri in #42120
feat(auto-router): add JEV classifier alongside LLM classifier by @devin-ai-integration[bot] in #41886
fix(proxy): park requeued spend logs in Redis so they survive a pod restart during a DB outage by @devin-ai-integration[bot] in #42022
fix(bedrock): forward anthropic-beta headers verbatim on the Claude platform messages path by @devin-ai-integration[bot] in #42275
chore(prices): sync AWS Bedrock prices: 4 models [enrichment failed: AWS Bedrock, 26 held] by @berriai-litellm-provider-info-sync[bot] in #42279
chore(prices): sync OpenRouter prices: 3 models by @berriai-litellm-provider-info-sync[bot] in #42280
fix(azure_ai): bridge gpt-5.4+ function tools with reasoning to the Foundry Responses API by @devin-ai-integration[bot] in #42041
fix: apply configured cache_control_injection_points beside client cache_control marks by @devin-ai-integration[bot] in #41956
fix(token_counter): count replayed redacted_thinking blocks so prompt_caching keeps pinning by @devin-ai-integration[bot] in #42069
fix(mcp): paginate prompt and resource discovery by @devin-ai-integration[bot] in #39189
chore(prices): sync AWS Bedrock prices: 3 models [enrichment failed: AWS Bedrock, 18 held] by @berriai-litellm-provider-info-sync[bot] in #42290
chore(prices): sync OpenRouter prices: 1 model by @berriai-litellm-provider-info-sync[bot] in #42289
fix(router): keep prompt caching affinity when the breakpoint moves by @devin-ai-integration[bot] in #42080
feat(ui): show prompt caching requests and net savings by @tin-berri in #42055
feat(bedrock_mantle): serve /v1/messages for Claude models on Mantle's native Anthropic Messages API by @devin-ai-integration[bot] in #42049
feat(policy_engine): add default fallback policy attachments by @devin-ai-integration[bot] in #42119
fix(mcp): handle split UTF-8 routing previews by @devin-ai-integration[bot] in #34919
fix(bedrock): send s3BucketOwner on batch input and output data config by @devin-ai-integration[bot] in #42262
feat(rust): scaffold cache foundation for Python parity by @devin-ai-integration[bot] in #42196
fix(fal_ai): price images from the dimensions fal returns by @devin-ai-integration[bot] in #42282
chore(prices): sync AWS Bedrock prices: 1 model [enrichment failed: AWS Bedrock, 16 held] by @berriai-litellm-provider-info-sync[bot] in #42298
chore(prices): sync OpenRouter prices: 1 model by @berriai-litellm-provider-info-sync[bot] in #42297
feat(fal_ai): add MiniMax H3 text-to-video and reference-to-video by @devin-ai-integration[bot] in #42286
fix(ci): let the install smoke test boot its key-less proxy config by @devin-ai-integration[bot] in #42296
fix: rename the mainland China brand to Qianwen AI Platform by @devin-ai-integration[bot] in #42284
test(integration): endpoint, breakdown component and failure support in the cost harness by @devin-ai-integration[bot] in #41999
test(integration): embeddings, rerank, completions and moderations cost cases by @devin-ai-integration[bot] in #42020
chore(prices): sync OpenRouter prices: 4 models, 3 new by @berriai-litellm-provider-info-sync[bot] in #42305
test(integration): audio, image and per-unit cost cases by @devin-ai-integration[bot] in #42024
feat(ui): add upgrade banner with latest release changelog stats by @kerry-berri in #40429
fix(fal_ai): surface fal errors in video status and content instead of completed and generic 500 by @devin-ai-integration[bot] in #42306
test(integration): passthrough route cost cases by @devin-ai-integration[bot] in #42028
fix(proxy): detach stored credential when model editor selects None by @devin-ai-integration[bot] in #42291
fix(proxy): renew budget reservation counter TTL while the request is in flight by @devin-ai-integration[bot] in #40322
chore(prices): sync OpenRouter prices: 1 model by @berriai-litellm-provider-info-sync[bot] in #42320
test(integration): pricing dimension and provider reported cost cases by @devin-ai-integration[bot] in #42035
fix(anthropic): forward Claude Code safeguards and dangerous-tool-use beta to Bedrock Invoke and Vertex on /v1/messages by @devin-ai-integration[bot] in #42288
feat(edenai): add Eden AI provider across chat, Responses, Messages, embeddings, audio, images and video by @hMED22 in #41101
feat(rust-cache): serve RedisClusterCache natively as a Redis topology by @devin-ai-integration[bot] in #42317
test(integration): provider wire cost cases by @devin-ai-integration[bot] in #42052
test(e2e): replace custom endpoints_client with provider SDK clients by @mateo-berri in #34358
fix(mcp): keep oauth scopes in admin api credential redaction by @devin-ai-integration[bot] in #39805
fix(otel v2): map OCR page markdown onto the generation output by @devin-ai-integration[bot] in #42267
test(integration): proxy behaviour cost cases by @devin-ai-integration[bot] in #42060
chore(prices): sync OpenRouter prices: 1 model by @berriai-litellm-provider-info-sync[bot] in #42333
test(integration): batch and realtime cost cases by @devin-ai-integration[bot] in #42066
fix(proxy): keep config-defined deployments when a config read returns no model_list by @devin-ai-integration[bot] in #41505
feat(rust): add CyberArk Conjur secret manager backend by @devin-ai-integration[bot] in #42303
fix(rust): preserve Python settings semantics at the native boundary by @devin-ai-integration[bot] in #42300
fix(alerting): deliver every distinct alert queued in one flush window by @devin-ai-integration[bot] in #42314
fix(bedrock): keep batch S3 credentials out of chat requests and debug logs by @devin-ai-integration[bot] in #42312
fix(proxy): evict the cached user row when SCIM or /user/delete removes a user by @devin-ai-integration[bot] in #42315
fix(auth): fail closed when the team membership lookup hits a db outage by @devin-ai-integration[bot] in #42036
test(migrations): cover the release-to-release upgrade path by @yuneng-berri in #42294
feat(rust): add Azure Key Vault secret manager backend by @devin-ai-integration[bot] in #42309
feat(rust): native Azure Blob response cache backend by @devin-ai-integration[bot] in #42321
chore(prices): sync AWS Bedrock prices: 6 models [enrichment failed: AWS Bedrock, 4 held] by @berriai-litellm-provider-info-sync[bot] in #42338
chore(prices): sync OpenRouter prices: 3 models by @berriai-litellm-provider-info-sync[bot] in #42337
refactor(types): remove 1,173 Any errors across 169 backend files by @mateo-berri in #40251
fix(responses): stream one lifecycle across MCP auto-execute rounds by @Atharva-Kanherkar in #40121
fix(responses): stop agentic follow-up from passing request params twice by @chopratejas in #41560
fix(streaming): keep an explicit provider prompt_tokens=0 or completion_tokens=0 in streamed usage by @devin-ai-integration[bot] in #42323
fix(router): walk every entry of a fallback list after a mid-stream failure by @devin-ai-integration[bot] in #42283
refactor(mcp): extract explicit operation context and dispatch by @joshua-berri in #42292
test: fix stale budget-status and bad-database-url assertions by @yuneng-berri in #42339
chore(prices): sync OpenRouter prices: 3 models by @berriai-litellm-provider-info-sync[bot] in #42349
fix(responses): patch custom_tool_call_output in place on guardrail write-back by @chopratejas in #41561
feat(rust): add native disk cache backend by @devin-ai-integration[bot] in #42311
feat(auth): breached password detection, self-service change-password and forced password reset by @devin-ai-integration[bot] in #42278
feat(docker): add a quickstart compose file served from the product repo by @yuneng-berri in #42326
fix(xai): accept max_completion_tokens as a supported param by @devin-ai-integration[bot] in #42353
feat(rust): add native GCS object-store cache backend by @devin-ai-integration[bot] in #42325
feat(rust): add HashiCorp Vault secret manager crate by @devin-ai-integration[bot] in #42308
fix(proxy): apply DB-stored callback redaction settings before logger init by @devin-ai-integration[bot] in #42122
fix(mcp): keep server lists stable across refreshes by @devin-ai-integration[bot] in #41074
chore(prices): sync OpenRouter prices: 4 models, 3 deprecated by @berriai-litellm-provider-info-sync[bot] in #42357
feat(rust): add native S3 cache backend by @devin-ai-integration[bot] in #42313
feat(xiaomi_mimo): add mimo-v2.6-pro and mimo-v2.6-flash cost map rows with live e2e coverage by @devin-ai-integration[bot] in #42362
fix(bedrock): add bare moonshotai.kimi-k3 cost map entry by @devin-ai-integration[bot] in #42363
feat(cache): add native Valkey semantic cache backend by @devin-ai-integration[bot] in #42316
fix(logging_worker): make flush() survive an event loop change by @devin-ai-integration[bot] in #42355
fix(auth): let jwt team_allowed_routes paths grant auth=true passthrough by @ryan-crabbe-berri in #42346
fix(rust): declare _CacheTestHandle.valkey_semantic in native stub by @devin-ai-integration[bot] in #42364
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42365
feat(ui): add internal-user savings and auto-router usage by @tin-berri in #42026
test(model_management): cover actor edges and wildcard models by @yuneng-berri in #41769
feat(ui): expose remaining complexity router advanced settings by @devin-ai-integration[bot] in #42293
test(ui): cover dashboard form journeys by @yuneng-berri in #41773
test(router): cover legacy lowest TPM selection by @yuneng-berri in #41795
feat(proxy): report the source of alerting, UI and router settings on read by @yuneng-berri in #41788
fix(proxy): record aborted outcome when spend-log cleanup is cancelled at shutdown by @yucheng-berri in #41213
fix(router): make context-window escalation opt-in by @tin-berri in #41872
feat(cache): serve Redis Semantic caches natively in Rust by @devin-ai-integration[bot] in #42319
fix(cli): preserve newer installed status lines during setup by @tin-berri in #42356
chore(prices): sync OpenRouter prices: 4 models by @berriai-litellm-provider-info-sync[bot] in #42371
feat(agents): attach access groups to agents and enforce them for models, MCP servers and agent calls by @devin-ai-integration[bot] in #41634
fix(guardrails): run key-attached guardrails on /v1/videos by @devin-ai-integration[bot] in #42354
fix(azure): propagate asyncio.CancelledError instead of raising a 500 by @devin-ai-integration[bot] in #42295
test(e2e): cover chat and responses registry gaps by @yuneng-berri in #41794
feat(cache): serve QdrantSemanticCache natively from Rust by @devin-ai-integration[bot] in #42324
chore(prices): sync OpenRouter prices: 4 models by @berriai-litellm-provider-info-sync[bot] in #42377
feat(ui): show Capability and FUSE v2 routing forecasts by @tin-berri in #42057
fix(cache): keep native Redis semantic binding and Qdrant batch writes after merge by @devin-ai-integration[bot] in #42379
feat(ui): add per-user breakdown to team usage export by @devin-ai-integration[bot] in #42367
feat(jwt-key-mapping): accept token_id as an alternative to the plaintext key by @Louis-Vauterin in #39578
fix(openrouter): remove the retired stealth/union-alpha model from the cost map by @devin-ai-integration[bot] in #42386
feat(pricing): add xai grok-4.20 aliases and image token prices from /v1/language-models by @devin-ai-integration[bot] in #42384
test(proxy): isolate the agent read-through singleton between unknown-agent tests by @devin-ai-integration[bot] in #42389
fix(mcp): return camelCase tool keys from /v1/mcp/tools after the SDK 2 upgrade by @devin-ai-integration[bot] in #42352
fix(logging): price terminal Responses stream events from their inner response by @devin-ai-integration[bot] in #42385
fix(bedrock): send every Mantle beta in the anthropic-beta header on the bedrock/mantle route by @devin-ai-integration[bot] in #42376
fix(bedrock): sign batch S3 requests with s3_access_key_id and s3_secret_access_key by @devin-ai-integration[bot] in #42342
feat(fal_ai): add flux-lora-depth image edits and moondream3 chat completions by @devin-ai-integration[bot] in #42334
feat(proxy): opt-in litellm_call_id in JSON error bodies by @devin-ai-integration[bot] in #42391
fix(fal_ai): price non-canonical image sizes from the nearest row and honour dump options by @devin-ai-integration[bot] in #42336
fix(proxy): drop cost-map metadata echoed back on model save by @devin-ai-integration[bot] in #41944
feat(openrouter): price typesafe/jev-1.13 and add an openrouter decisions pass-through by @devin-ai-integration[bot] in #42301
feat(cost): warn and count $0 cost on billable requests by @devin-ai-integration[bot] in #42345
fix(proxy): release unclaimed budget reservations at request end by @devin-ai-integration[bot] in #42304
test(integration): move Xiaomi MiMo coverage from live e2e to the providers wire shard by @devin-ai-integration[bot] in #42395
feat(fal_ai): add queue-only /fal_ai pass-through route with spend tracking by @devin-ai-integration[bot] in #42360
fix(auth): fail closed when the JWT single-team fallback or compact editor membership read hits a DB outage by @devin-ai-integration[bot] in #42344
refactor(agentic-loop): build follow-up kwargs in one place so no executor can repeat a request param by @ryan-crabbe-berri in #42307
fix(responses): drop client_metadata and merge system messages for Databricks chat-only models by @devin-ai-integration[bot] in #42390
test(e2e-ui): check the MCP Tools tab against the upstream's own tools/list by @devin-ai-integration[bot] in #42397
test(integration): chain a proxy-issued previous_response_id in the cost suite by @devin-ai-integration[bot] in #42396
feat(rust-bridge): add cache and secret migration foundations by @devin-ai-integration[bot] in #42328
fix(proxy): surface a database outage from the user read as 503 no_db_connection by @devin-ai-integration[bot] in #42399
test(pricing): assert cache-priced vertex grok rows advertise supports_prompt_caching by @kumarpriyanshu09 in #41526
fix(bedrock): price bedrock/mantle/ deployments from the base model row by @devin-ai-integration[bot] in #42402
fix(anthropic_adapter): keep reasoning_effort a string for targets that stay on chat completions by @devin-ai-integration[bot] in #42401
chore(prices): sync OpenRouter prices: 2 models, 2 deprecated by @berriai-litellm-provider-info-sync[bot] in #42381
feat(proxy): add TinyFish Agent API passthrough with per-step billing by @Zechereh in #41099
chore(prices): sync OpenRouter prices: 2 models, 2 new by @berriai-litellm-provider-info-sync[bot] in #42407
ci: skip cost map file checks on PRs that leave the cost map untouched by @devin-ai-integration[bot] in #42406
feat(tokenizer): preserve Python defaults with opt-in Rust dispatch by @devin-ai-integration[bot] in #42174
fix(presidio): mask PII in streaming /v1/messages output by @devin-ai-integration[bot] in #42351
fix(otel v2): map completions, images, speech, transcription and moderation output onto the Langfuse generation output by @devin-ai-integration[bot] in #42394
fix(cost): honor per-second custom pricing on chat completions for every provider by @devin-ai-integration[bot] in #42403
chore(prices): sync OpenRouter prices: 2 models by @berriai-litellm-provider-info-sync[bot] in #42418
feat(errors): prefilled GitHub issue link on unmapped internal errors by @devin-ai-integration[bot] in #42065
test(e2e): add conversational matrix across chat, messages and responses by @devin-ai-integration[bot] in #42359
fix(proxy): answer 503 no_db_connection on management routes when the caller's user read hits a database outage by @devin-ai-integration[bot] in #42410
fix(e2e-stack): print add-mask lines only under GitHub Actions by @devin-ai-integration[bot] in #42423
feat(errors): add stream and safe config flags to the bug report link by @ryan-crabbe-berri in #42428
feat(router): native compact-to-fit across conversation APIs by @tin-berri in #42074
fix(mcp): keep config-defined servers read-only by @joshua-berri in #42299
ci(code-quality): allowlist _render_json in the recursive detector by @devin-ai-integration[bot] in #42442
fix(guardrails): store the masked output in spend logs when Presidio masks the response by @devin-ai-integration[bot] in #42441
fix(jwt): accept a team alias in x-litellm-team-id by @devin-ai-integration[bot] in #42445
chore(prices): sync OpenRouter prices: 14 models, 1 deprecated [enrichment failed: OpenRouter, 5 held] by @berriai-litellm-provider-info-sync[bot] in #42438
feat(rust): align secret manager operation contexts by @devin-ai-integration[bot] in #42480
chore(prices): sync OpenRouter prices: 1 model [enrichment failed: OpenRouter, 5 held] by @berriai-litellm-provider-info-sync[bot] in #42485
feat(arize): per-team success and error sampling rates for the Arize AX callback by @devin-ai-integration[bot] in #42383
test(unit): make bedrock collector and secret scan timing tests deterministic by @devin-ai-integration[bot] in #42405
feat(anthropic): add Claude Opus 5.5 by @devin-ai-integration[bot] in #42489
fix(gemini): simplify model version check by @philschmid in #42465
fix(bedrock/claude_platform): strip body params the AWS endpoint rejects by @mateo-berri in #31203
fix(proxy): never render credential-bearing config keys in the bug report by @devin-ai-integration[bot] in #42493
fix(cost): bill batch prompts above 272K at OpenAI's long-context batch tier by @mateo-berri in #39861
fix(otel v2): map rerank and search output and the OCR, image edit and search input onto the Langfuse generation by @devin-ai-integration[bot] in #42444
test(response_metadata): make the detailed-timing receive-anchor test timezone independent by @devin-ai-integration[bot] in #42429
fix(utils): stop a nested additional_drop_params entry from crashing openai-compatible calls by @yuneng-berri in #42492
fix(jwt): say x-litellm-team-id matched no team id or alias in the 403 by @devin-ai-integration[bot] in #42495
fix(ui): surface the owner's user budget on keys without their own budget by @devin-ai-integration[bot] in #38220
test(proxy): give every ui settings endpoint test a fresh settings store by @devin-ai-integration[bot] in #42430
chore(prices): sync Azure prices: 34 models by @berriai-litellm-provider-info-sync[bot] in #42502
chore(prices): sync OpenAI prices: 25 models [enrichment failed: OpenAI, 32 held] by @berriai-litellm-provider-info-sync[bot] in #42501
feat(router): time-windowed team reservation of deployments via model_info.access_windows by @devin-ai-integration[bot] in #42398
feat(terraform): expose server_metadata on litellm_key so undeclared metadata is visible by @devin-ai-integration[bot] in #42453
fix(mcp): restore legacy SSE and bounded cancellation cleanup by @joshua-berri in #42382
chore(prices): sync Google Gemini prices: 13 models by @berriai-litellm-provider-info-sync[bot] in #42500
fix(realtime): surface an upstream handshake refusal as an error event and policy close by @devin-ai-integration[bot] in #42388
feat(openai): add GPT-6 Sol and GPT-6 Luna by @devin-ai-integration[bot] in #42515
fix(router): explain fallback outcome in plain words in the raised error by @devin-ai-integration[bot] in #42509
fix(cost): honor deployment pricing for image generation by @patel-26meet in #39311
fix(streaming): let a later usage event zero out stale cache counts (#40736) by @devin-ai-integration[bot] in #42330
fix(fal_ai): handle seconds=auto and oversized sizes for minimax h3 videos by @devin-ai-integration[bot] in #42504
fix(proxy): write key deleted audit logs for cascade and alias key deletions by @devin-ai-integration[bot] in #42446
fix(rust_bridge): keep Messages, token counter and tokenizer routes on Python by @devin-ai-integration[bot] in #42517
chore(pricing): remove retired models flagged by the provider sync by @devin-ai-integration[bot] in #42521
feat(proxy): admin-only /debug/report sharing the bug report environment by @devin-ai-integration[bot] in #42440
fix(aws_secret_manager_v2): restore secret scheduled for deletion instead of failing CreateSecret by @devin-ai-integration[bot] in #42454
feat(rust): add python-compat crate for Python data formats by @devin-ai-integration[bot] in #42510
fix(langsmith): json.dumps with default=str so non-serializable metadata does not crash batch flush by @devin-ai-integration[bot] in #42424
fix(proxy): make the lazy OpenAPI snapshot byte-identical on every Python version by @devin-ai-integration[bot] in #42519
test(rust_bridge): drop route dispatch assertions, test the bridge directly by @devin-ai-integration[bot] in #42536
fix(websearch_interception): keep intercepted searches under the parent request's session and trace by @devin-ai-integration[bot] in #41711
test(e2e): assert a cooldown reaches a sibling replica within the 1s Redis read interval by @devin-ai-integration[bot] in #42422
feat(proxy): opt-in include_guardrail_response returns guardrail_information in the response by @devin-ai-integration[bot] in #42327
feat(logs): add span type filter to request logs by @devin-ai-integration[bot] in #42491
fix(e2e): route credential, cost map, and UI login calls to the control plane by @devin-ai-integration[bot] in #42506
test(google_genai): move unified_google_tests to gemini-3.5-flash-lite by @devin-ai-integration[bot] in #42520
feat(router): add group-scoped priority routing strategy by @tin-berri in #42378
fix(fal_ai): align /fal_ai queue gate with the pricer and normalise resolution type by @devin-ai-integration[bot] in #42505
fix(fal_ai): reuse the status client and headers on the video result probe by @devin-ai-integration[bot] in #42511
refactor(rust): align the cache crates with Python and wire every native backend by @devin-ai-integration[bot] in #42530
fix(fal_ai): honour global api_base for image generation and reject non-string reasoning_effort with 400 by @devin-ai-integration[bot] in #42512
fix(ui): rename reminder markers to Ignore Custom Tags by @tin-berri in #42370
chore(prices): sync Baseten prices: 12 models, 9 new [enrichment failed: Baseten, 15 held] by @berriai-litellm-provider-info-sync[bot] in https://github.com/BerriAI/litellm/pull/42558
fix: enforce disable_custom_api_keys from general_settings by @yuneng-berri in https://github.com/BerriAI/litellm/pull/42437
feat: add configurable provider affinity header mapping by @togear in https://github.com/BerriAI/litellm/pull/41033
fix(proxy): share model rate-limit buckets between a model_group_alias and its target by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42516
test(e2e): one request lands the same spend on every surface by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42540
fix(mcp): apply post-call rewrites without stale structured output by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/41530
chore(prices): sync Baseten prices: 4 models, 4 new [enrichment failed: Baseten, 3 held] by @berriai-litellm-provider-info-sync[bot] in https://github.com/BerriAI/litellm/pull/42578
fix(otel): honor SSL_CERT_FILE and ssl_verify in OTLP HTTP exporters by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42106
chore(prices): sync OpenAI prices: 3 models by @berriai-litellm-provider-info-sync[bot] in https://github.com/BerriAI/litellm/pull/42557
fix(ssrf): point the blocked-address remediation at litellm_settings by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42508
fix(model_prices): registry audit 2026-09-22, absorb open pricing PRs by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42543
fix(gateway): expose /api/event_logging/batch on the gateway allowlist by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42572
fix(proxy): honor DATABASE_DISABLE_PREPARED_STATEMENTS in the litellm CLI by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42556
chore(prices): sync Vertex AI prices: 20 models by @berriai-litellm-provider-info-sync[bot] in https://github.com/BerriAI/litellm/pull/42577
fix(proxy): attribute provider and model_info on pre_call_hook rejections by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/41077
chore(cost-map): remove models past their deprecation date by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42435
test(vcr): guard leaked cassette patches and make injected-transport embedding tests immune by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42542
fix(otel): keep text completion choice fields beside the synthesized message by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42537
fix(mcp): preserve discovery attribution and sanitize logging headers by @joshua-berri in https://github.com/BerriAI/litellm/pull/42541
fix(proxy_cli): import proxy_server once on script-style boot by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42584
feat(bedrock): add Claude Opus 5.5 pricing and capabilities by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42588
test(e2e): hold every worker under an idle RSS budget before any traffic by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42552
feat(cost-map): add Claude Opus 5.5 for Vertex AI and Azure AI by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42599
feat(proxy): configurable key_alias_pattern for key generate, update, and regenerate by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42553
test(proxy): make two proxy-infra tests independent of sibling-test state by @yuneng-berri in https://github.com/BerriAI/litellm/pull/42581
test: count a zombie grandchild as gone in the migrate deploy timeout test by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42570
test(e2e): run the memory cell alone on the shared stack by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42518
fix(caching): keep embedding cache hits aligned with request inputs by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42571
chore(prices): sync xAI prices: 3 models, 3 new [3 with gaps] by @berriai-litellm-provider-info-sync[bot] in https://github.com/BerriAI/litellm/pull/42591
fix(otel): record the GenAI exception event through the Logs API on both OpenTelemetry lines by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42431
fix(spend): return 400 from /spend/calculate for a model with no pricing row by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42497
fix(anthropic): return 400 instead of 500 when a content list holds a bare string by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42420
feat(rust): add standalone cost calculator by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42604
feat(logging): add normalized_error cluster key to error_information by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/41715
test(bedrock): point unit tests at model ids still in the cost map by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42606
feat(rust): add immutable model catalog crate by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42605
fix(ui): let the Create Key user picker find users by user_id, not just email by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/41687
test(cost_calculator): point image-generation deployment price test at a live gemini row by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42615
test(utils): accept the per-size image cost keys in the price-map schema check by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42612
fix(ollama): send PNG and JPEG images without requiring Pillow. by @Pawan-Shahane in https://github.com/BerriAI/litellm/pull/41979
fix: answer get_api_base for github_copilot and chatgpt without running the login flow by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42602
test: point CircleCI-only suites at models still in the cost map by @yuneng-berri in https://github.com/BerriAI/litellm/pull/42617
fix: drop duplicate cache_read_input_token_cost_batches keys from the price maps by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42623
fix(s3): replace colons in generated log filenames by @mubashir1osmani in https://github.com/BerriAI/litellm/pull/40452
test(realtime): drop legacy InvalidStatusCode tests and pin websockets imports by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42624
fix(bedrock): treat blank AWS_S3_* env vars as unset for batch jobs by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42528
ci(test-unit): drop dead misc shard paths and skip missing paths with a warning by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42603
fix(proxy): keep the in-flight daily spend batch when shutdown cancels the flush by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42593
fix(utils): isolate callback errors in async_post_call_success_deployment_hook by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42535
test(e2e): add secret manager lanes for HashiCorp Vault and CyberArk Conjur by @yujonglee-berri in https://github.com/BerriAI/litellm/pull/42503
feat(ui): simplify auto-router setup and clarify feature limits by @moe-berri in https://github.com/BerriAI/litellm/pull/42625
feat(ui): add native LiteAdmin assistant by @tin-berri in https://github.com/BerriAI/litellm/pull/42443
fix(mcp): preserve credential authority in DCR bridge authentication by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42563
feat(ci): close open pull requests superseded by a merged fix on their linked issue by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42609
test(cost): repoint the Azure image cost test at gpt-image-2 by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42631
feat(logger): dispatch Python logging through the Rust diagnostics processor by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42616
fix(bedrock): drop unsupported sampling params on converse reasoning models by @emerzon in https://github.com/BerriAI/litellm/pull/39834
test(utils): raise the post-success hook error from a guardrail in the failure-hook regression by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/42646
chore(docker): bump wolfi-base digest to pick up glibc 2.44-r6 by @yuneng-berri in https://github.com/BerriAI/litellm/pull/42643
chore: bump litellm-enterprise 0.1.69 -> 0.1.70, litellm-proxy-extras 0.4.100 -> 0.4.101, litellm 1.103.0 -> 1.104.0 by @yuneng-berri in https://github.com/BerriAI/litellm/pull/42633
fix: repair seven regressions caught by CircleCI on main by @yuneng-berri in https://github.com/BerriAI/litellm/pull/42640

New Contributors

@clonylu made their first contribution in #41203
@zoroyihan7 made their first contribution in #40243
@adssoccer1 made their first contribution in #40399
@gaurav-pandey-zocdoc made their first contribution in #39102
@abhirup7 made their first contribution in #40147
@chopratejas made their first contribution in #41560
@kumarpriyanshu09 made their first contribution in #41526
@Zechereh made their first contribution in #41099
@philschmid made their first contribution in #42465
@patel-26meet made their first contribution in #39311
@togear made their first contribution in https://github.com/BerriAI/litellm/pull/41033
@Pawan-Shahane made their first contribution in https://github.com/BerriAI/litellm/pull/41979

Full Changelog: v1.103.0-dev.2...v1.104.0-dev.1