* fix(triton/completion/transformation.py): remove bad_words / stop words from triton call
parameter 'bad_words' has invalid type. It should be either 'int', 'bool', or 'string'.
* fix(proxy_track_cost_callback.py): add debug logging for track cost callback error
* feat(fireworks_ai/chat): handle tool calling with fireworks ai correctly
Fixes https://github.com/BerriAI/litellm/issues/7209
* fix(utils.py): handle none type in message
* fix: fix model name in test
* fix(utils.py): fix validate check for openai messages
* fix: fix model returned
* fix(main.py): fix text completion routing
* test: update testing
* test: skip test - cohere having RBAC issues
* build(model_prices_and_context_window.json): add vertex ai gemini-2.5-flash pricing
* build(model_prices_and_context_window.json): add gemini reasoning token pricing
* fix(vertex_and_google_ai_studio_gemini.py): support counting thinking tokens for gemini
allows accurate cost calc
* fix(utils.py): add reasoning token cost calc to generic cost calc
ensures gemini-2.5-flash cost calculation is accurate
* build(model_prices_and_context_window.json): mark gemini-2.5-flash as 'supports_reasoning'
* feat(gemini/): support 'thinking' + 'reasoning_effort' params + new unit tests
allow controlling thinking effort for gemini-2.5-flash models
* test: update unit testing
* feat(vertex_and_google_ai_studio_gemini.py): return reasoning content if given in gemini response
* test: update model name
* fix: fix ruff check
* test(test_spend_management_endpoints.py): update tests to be less sensitive to new keys / updates to usage object
* fix(vertex_and_google_ai_studio_gemini.py): fix translation
* fix(proxy/_types.py): allow internal user to call api playground
* fix(new_usage.tsx): cleanup tag based usage - only show for proxy admin
not clear what tags internal user should be allowed to see
* fix(team_endpoints.py): allow internal user view spend for teams they belong to
* fix(team_endpoints.py): return team alias on `/team/daily/activity` API
allows displaying team alias on ui
* fix: fix linting error
* fix(entity_usage.tsx): allow viewing top keys by team
* fix(entity_usage.tsx): show alias, if available in breakdown
allows entity alias to be easily displayed
* Show usage by key (on all up, team, and tag usage dashboards) (#10152)
* fix(entity_usage.tsx): allow user to select team in team usage tab
* fix(new_usage.tsx): load all tags for filtering
* fix(tag_management_endpoints.py): return dynamic tags from db on `/tag/list`
* fix(litellm_pre_call_utils.py): support x-litellm-tags even if tag based routing not enabled
* fix(new_usage.tsx): show breakdown of usage by api key on dashboard
helpful when looking at spend by team
* fix(networking.tsx): exclude litellm-dashboard team id's from calls
adds noisy ui tokens to key activity
* fix(new_usage.tsx): allow user to see activity by key on main tab
* feat(internal_user_endpoints.py): refactor to use common_daily_activity function
reuses same logic across teams/keys/tags
Allows returning team_alias in api_keys consistently
* fix(leftnav.tsx): swap old usage with new usage tab
* fix(entity_usage.tsx): show breakdown of teams in daily spend chart
* style(new_usage.tsx): show global usage tab if user is admin / has admin view
* fix(new_usage.tsx): add disclaimer for new usage dashboard
* fix(new_usage.tsx): fix linting error
* Allow filtering usage dashboard by team + tag (#10150)
* fix(entity_usage.tsx): allow user to select team in team usage tab
* fix(new_usage.tsx): load all tags for filtering
* fix(tag_management_endpoints.py): return dynamic tags from db on `/tag/list`
* fix(litellm_pre_call_utils.py): support x-litellm-tags even if tag based routing not enabled
* fix: fix linting error
* initial commit for azure responses api support
* update get complete url
* fixes for responses API
* working azure responses API
* working responses API
* test suite for responses API
* azure responses API test suite
* fix test with complete url
* fix test refactor
* test fix metadata checks
* fix code quality check
* feat(llm_passthrough_endpoints.py): expose new `/vertex_ai/discovery/` endpoint
Allows calling vertex ai discovery endpoints via passthrough
For agentbuilder api calls
* refactor(llm_passthrough_endpoints.py): use common _base_vertex_proxy_route
Prevents duplicate code
* feat(llm_passthrough_endpoints.py): add vertex endpoint specific passthrough handlers
* feat(ui/): add team based usage to dashboard
allows admin to see spend across teams + within teams at 1m+ spend logs
* fix(entity_usage.tsx): add activity page to entity usage
* style(entity_usage.tsx): place filter above tab switcher
* feat: initial commit adding daily tag spend table to db
* feat(db_spend_update_writer.py): correctly log tag spend transactions
* build(schema.prisma): add new tag table to root
* build: add new migration file
* feat(common_daily_activity.py): add `/tag/daily/activity` API endpoint
allows viewing daily spend by tag
* feat(tag_management_endpoints.py): support comma separated list of tags + tag breakdown metric
allows querying multiple tags + knowing what tags are driving spend
* feat(entity_usage.tsx): initial commit adding tag based usage to litellm dashboard
brings back tag based usage tracking to UI at 1m+ spend logs
* feat(entity_usage.tsx): add top api key view to ui
* feat(entity_usage.tsx): add tag table to ui
* feat(entity_usage.tsx): allow filtering by tag
* refactor(entity_usage.tsx): reorder components
* build(ui/): fix linting error
* fix: fix ruff checks
* fix(schema.prisma): drop uniqueness requirement on tag
allows dailytagspend to have multiple rows with the same tag
* build(schema.prisma): drop uniqueness requirement on tag in dailytagspend
allows tag agg. view to work on multiple rows with same tag
* build(schema.prisma): drop tag uniqueness requirement
* feat: initial commit adding daily tag spend table to db
* feat(db_spend_update_writer.py): correctly log tag spend transactions
* build(schema.prisma): add new tag table to root
* build: add new migration file
* fix(llm_http_handler.py): fix fake streaming
allows groq to work with llm_http_handler
* fix(groq.py): migrate groq to openai like config
ensures json mode handling works correctly
* Simple fix for #9339 - upgrade the underlying library and cache the azure storage client (#9965)
* fix - use constants for caching azure storage client
---------
Co-authored-by: Adrian Lyjak <adrian@chatmeter.com>
* Add litellm_call_id passing to aim guardrails on pre and post-hooks
* Add test that ensures that pre_call_hook receives litellm call id when common_request_processing called
* fix(pub_sub.py): fix passing project id in pub sub call
Fixes issue where GCS_PUBSUB_PROJECT_ID was not being used
* test(test_pub_sub.py): add unit test to prevent future regressions
* test: fix test
* feat(schema.prisma): initial commit adding aggregate table for team spend
allows team spend to be visible at 1m+ logs
* feat(db_spend_update_writer.py): support logging aggregate team spend
allows usage dashboard to work at 1m+ logs
* feat(litellm-proxy-extras/): add new migration file
* fix(db_spend_update_writer.py): fix return type
* build: bump requirements
* fix: fix ruff error
