* Add async_post_call_response_headers_hook to CustomLogger (#20070)
Allow CustomLogger callbacks to inject custom HTTP response headers
into streaming, non-streaming, and failure responses via a new
async_post_call_response_headers_hook method.
* async_post_call_response_headers_hook
---------
Co-authored-by: michelligabriele <gabriele.michelli@icloud.com>
* fix(proxy): prevent provider-prefixed model leaks
Proxy clients should not see LiteLLM internal provider prefixes (e.g. hosted_vllm/...) in the OpenAI-compatible response model field.
This patch sanitizes the client-facing model name for both:
- Non-streaming responses returned from base_process_llm_request
- Streaming SSE chunks emitted by async_data_generator
Adds regression tests covering vLLM-style hosted_vllm routing for both streaming and non-streaming paths.
* chore(lint): suppress PLR0915 in proxy handler
Ruff started flagging ProxyBaseLLMRequestProcessing.base_process_llm_request() for too many statements after the hotpatch changes.
Add an explicit '# noqa: PLR0915' on the function definition to avoid a large refactor in a hotpatch.
* refactor(proxy): make model restamp explicit
Replace silent try/except/pass and type ignores with explicit model restamping.
- Logs an error when the downstream response model differs from the client-requested model
- Overwrites the OpenAI `model` field to the client-requested value to avoid leaking internal provider-prefixed identifiers
- Applies the same behavior to streaming chunks, logging the mismatch only once per stream
* chore(lint): drop PLR0915 suppression
The model restamping bugfix made `base_process_llm_request()` slightly exceed Ruff's
PLR0915 (too-many-statements) threshold, requiring a `# noqa` suppression.
Collapse consecutive `hidden_params` extractions into tuple unpacking so the
function falls back under the lint limit and remove the suppression.
No functional change intended; this keeps the proxy model-field bugfix intact
while aligning with project linting rules.
* chore(proxy): log model mismatches as warnings
These model-restamping logs are intentionally verbose: a mismatch is a useful signal
that an internal provider/deployment identifier may be leaking into the public
OpenAI response `model` field.
- Downgrade model mismatch logs from error -> warning
- Keep error logs only for cases where the proxy cannot read/override the model
* fix(proxy): preserve client model for streaming aliasing
Pre-call processing can rewrite request_data['model'] via model alias maps.\n\nOur streaming SSE generator was using the rewritten value when restamping chunk.model, which caused the public 'model' field to differ between streaming and non-streaming responses for alias-based requests.\n\nStash the original client model in request_data as _litellm_client_requested_model after the model has been routed, and prefer it when overriding the outgoing chunk model. Add a regression test for the alias-mapping case.
* chore(lint): satisfy PLR0915 in streaming generator
Ruff started flagging async_data_generator() for too many statements after adding model restamping logic.\n\nExtract the client-model selection + chunk restamping into small helpers to keep behavior unchanged while meeting the project's PLR0915 threshold.
The /health/services endpoint rejected datadog_llm_observability as an
unknown service, even though it was registered in the core callback
registry and __init__.py. Added it to both the Literal type hint and
the hardcoded validation list in the health endpoint.
* fix(vertex_ai): replace custom model names with actual Vertex AI model names in passthrough URLs (#19948)
When the passthrough URL already contains project and location, the code
was skipping the deployment lookup and forwarding the URL as-is to Vertex AI.
For custom model names like gcp/google/gemini-2.5-flash, Vertex AI returned
404 because it only knows the actual model name (gemini-2.5-flash).
The fix makes the deployment lookup always run, so the custom model name
gets replaced with the actual Vertex AI model name before forwarding.
* add _resolve_vertex_model_from_router
* fix: get_llm_provider
* Potential fix for code scanning alert no. 4020: Clear-text logging of sensitive information
Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
---------
Co-authored-by: michelligabriele <gabriele.michelli@icloud.com>
Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
* Cleanup code for user cli auth, and make sure not to prompt user for team multiple times while polling
* Adding tests
* Cleanup normalize teams some more
* Fix PLR0915: Extract system message handling to reduce statement count
* fix mypy
* fix: add host_progress_callback parameter to mock_call_tool in test
The test_call_tool_without_broken_pipe_error was failing because the mock function did not accept the host_progress_callback keyword argument that the actual implementation passes to client.call_tool(). Updated the mock to accept this parameter to match the real implementation signature.
* fixing flaky tests around oidc and email
* Add documentation comment to test file
* add retry
* add dependency
* increase retry
---------
Co-authored-by: yuneng-jiang <yuneng.jiang@gmail.com>
- Added a Pydantic validator to convert empty string inputs for max_budget to None, preventing float parsing errors from the frontend.
- Modified the internal user update logic to explicitly allow max_budget to be None, ensuring the value isn't filtered out and can be reset to unlimited in the database.
- Added unit tests for validation and logic.
Closes#19781
* fix(proxy): use return value from CustomLogger.async_post_call_success_hook
Previously the return value was ignored for CustomLogger callbacks,
preventing users from modifying responses. Now the return value is
captured and used to replace the response (if not None), consistent
with CustomGuardrail and streaming iterator hook behavior.
Fixes issue with custom_callbacks not being able to inject data into
LLM responses.
* fix(proxy): also fix async_post_call_streaming_hook to use return value
Previously the streaming hook only used return values that started with
"data: " (SSE format). Now any non-None return value is used, consistent
with async_post_call_success_hook and streaming iterator hook behavior.
Added tests for streaming hook transformation.
---------
Co-authored-by: Gabriele Michelli <michelligabriele0@gmail.com>
* fix(proxy): support slashes in google route params
* fix(proxy): extract google model ids with slashes
* test(proxy): cover google model ids with slashes
* [Feat] Add model parameter to Generic Guardrail API
Add model information to guardrail requests, allowing guardrails to make
model-specific security decisions.
Changes:
- Add `model` field to GenericGuardrailAPIInputs TypedDict
- Add `model` field to GenericGuardrailAPIRequest Pydantic model
- Update OpenAI and Anthropic handlers to pass model from request/response
- Add unit tests for model parameter handling
* [Feat] Add model parameter to all guardrail_translation handlers
Extend model parameter support to all guardrail handlers for consistent
implementation across all endpoint types:
- OpenAI Responses API (input/output + streaming)
- OpenAI Image Generation (input only)
- OpenAI Text Completion (input/output)
- OpenAI Text-to-Speech (input only)
- OpenAI Audio Transcription (output only)
- Cohere Rerank (input only)
- Pass-through Endpoints (input/output)
- MCP Server (input only)
This addresses the review feedback requesting consistent model parameter
handling across all guardrail_translation/handler.py files.
---------
Co-authored-by: Igal Boxerman <igal@pillar.security>
Ishaan Jaff