* Added type casting for messages to enhance type safety.
* Updated the handling of system messages to ensure correct content formatting.
* Cleaned up code for better readability and maintainability.
* Added support for reasoning parameters in magistral models, including "reasoning_effort" and "thinking".
* Updated the MistralConfig class to handle reasoning system prompts.
* Implemented tests to verify reasoning functionality and ensure correct parameter mapping for magistral models.
* Enhanced the model prices JSON to reflect new reasoning capabilities.
* Checkpoint before follow-up message
* Add comprehensive tests for Deepgram transcription functionality
* clean up transform
* just use 1 test
* test cleanup
* test fix get_complete_url
* test rename file
* refactor deepgram URL construction
* add logging_obj.pre_call
* fix unused imports
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
* Add Deepgram provider to supported providers list and mappings
* add logo
* Add deepgram to model cost map
* ui - require api key for deepgram
* fix logo path
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
* fix: passing litellm_params_dict instead of litellm_params which is a list of strings
* updated custom llm embedding params and litellm_params_dict
---------
Co-authored-by: Ganesh Jagadeesan <cjagadeesan@dataminr.com>
* fix(internal_user_endpoints.py): support user with `+` in email on user info
ensures user is correctly parsed from input
* fix(factory.py): support vertex function call args as None
handles empty string in args for vertex gemini calls
* docs(langfuse_integration.md): pin langfuse sdk version on docs
* fix(vertex_ai/): return empty dict, instead of none when empty string given
* refactor: reduce function size
* fix: fix linting errors
* fix: revert check
* fix(internal_user_endpoints.py): fix check
* test: update tests
* test: update tests
* fix(vertex_ai/anthropic): extract vertex credentials from correct dict
Fixes passing credentials for vertex anthropic via `/v1/messages`
* fix(vertex_ai/anthropic): working e2e vertex ai anthropic messages via `/v1/messages`
Ensures vertex anthropic models can be called via /v1/messages
* fix(vertex_passthrough_logging_handler.py): fix vertex anthropic logging to emit the model name on spend logs
enables observability of passthrough route
* fix(vertex_passthrough_logging_handler.py): don't redefine variable
* fix(vertex_passthrough_logging_handler.py): log cost + token usage for anthropic passthrough requests on vertex ai
* fix(vertex_ai/anthropic): extract vertex credentials from correct dict
Fixes passing credentials for vertex anthropic via `/v1/messages`
* fix(vertex_ai/anthropic): working e2e vertex ai anthropic messages via `/v1/messages`
Ensures vertex anthropic models can be called via /v1/messages
* fix(vertex_passthrough_logging_handler.py): fix vertex anthropic logging to emit the model name on spend logs
enables observability of passthrough route
* fix(vertex_passthrough_logging_handler.py): don't redefine variable
* build(model_prices_and_context_window.json): add o3-pro pricing
* build(model_prices_and_context_window.json): add updated o3 model pricing
* build(model_prices_and_context_window.json): add new o3-pro model version
* chore(pangea-guardrail): Fix typo in debug message.
* docs(pangea-guardrail): Fix YAML example in pangea.md (README)."
* docs(pangea-guardrail): Update pangea.md (README).
* chore(pangea-guardrail): Format with Black.
* build(model_prices_and_context_window.json): add 'claude-opus-4' on vertexai (no @)
* build(model_prices_and_context_window.json): add claude sonnet 4 without 2
*@
* feat(internal_user_endpoints.py): assign user to orgs on user creation
allows user to be a member of orgs on creation - work to enable default orgs on UI
* fix(internal_user_endpoints.py): fix http_request
* docs(deploy.md): move docker recommendation to `main-stable`
* feat(enterprise/internal_user_endpoints.py): expose endpoint for checking available premium users
* feat(usage_indictor.tsx): add new element to help track remaining premium users
* feat(usage_indicator.tsx): show premium user remaining usage
allows users with user caps to know how much is left
* fix(vertex_and_google_ai_studio_gemini.py): bubble up stream is not finished, even if stop reason is given
prevents early completion of stream
Closes https://github.com/BerriAI/litellm/issues/11549
* fix(streaming_handler.py): respect is_finished = False in hidden params
internal logic for preventing ending stream early
* fix(litellm_license.py): add function to check if user is over limit
* fix(internal_user_endpoints.py): add function to check if user is over limit
* refactor: move test
* docs(customer_endpoints.py): document new param
* Feature/lasso guardrail (#9002)
* first version of lasso guardrail in litellm
* update to the new Lasso API
* change prod api_base and kill the request when lasso detect issue.
* change test for now api, local test pass
* add async tests
* all tests pass
* add docs for the new lasso guardrail
* Remove support for modes other than pre_call in Lasso guardrail
* code structure and naming
* only pre_call docs
* fix lint errors
* move test to the new location follows the same directory structure as litellm/.
* add lasso guard
* docs lasso docs
* add lasso guardrail
* fix lasso guardrail
---------
Co-authored-by: oroxenberg <oro@lasso.security>
* Update web_search.md to include new supported providers and models, enhance web search options, and improve documentation for using web search with various AI models.
* Update LiteLLM version in web_search.md to reflect the latest stable release.
* Fix formatting in web_search.md for model declaration consistency.
* docs(web_search.md): add configuration options for web search in config.yaml
This update introduces sections for setting default and custom web search options in the proxy config file, including examples for different models and context sizes. A note clarifies that users can override these settings in API requests.
