image_edit was not forwarding model_info/metadata to the logging object,
so custom_pricing was never detected. After PR #20679 stripped custom
pricing fields from the shared backend key, image_edit cost became 0.
Fixes#22244
Remove /v1/chat/completions and /v1/responses from supported_endpoints
and revert the audio model detection change since gpt-realtime-1.5 does
not go through Chat Completions.
gpt-realtime-1.5 supports Chat Completions with audio params but
is_model_gpt_audio_model only checked for "audio" in the model name.
Add "realtime" check so the audio parameter is passed through correctly.
New OpenAI realtime model released 2026-02-23. Adds pricing and capability
metadata for gpt-realtime-1.5 (32K context, 4K output, audio/image/text I/O).
Unlike gpt-realtime, this model also supports Chat Completions and Responses
endpoints (not just WebSocket).
Closes#22266
In a multi-worker Uvicorn setup, a client that reconnects to a different
worker sends an mcp-session-id that the new worker has never seen. The
MCP SDK returns 400 because the session is unknown.
Fix: add _handle_stale_mcp_session() which inspects the inbound
mcp-session-id header before the request reaches the SDK. If the
session is not in this worker's _server_instances:
- Non-DELETE: strip the header so the SDK creates a fresh session
- DELETE: return 200 immediately (idempotent, session already gone)
No new dependencies, no Redis, no latency added to the hot path.
Fixes https://github.com/BerriAI/litellm/issues/20992
* [Docs] Enable local file support
Implemented internal handling for converting file-type documents to the required format for OCR processing, ensuring seamless integration with various providers.
* Refactor OCR file handling and improve security checks
Removed deprecated MIME type mapping and file conversion functions, replacing them with updated implementations. Enhanced security by rejecting 'file' document types in JSON requests, ensuring file uploads are handled via multipart/form-data. Updated tests to reflect these changes and ensure proper functionality.
* Enhance MIME type validation in OCR processing
Added a regular expression check to validate MIME types in the convert_file_document_to_url_document function, raising a ValueError for invalid types. Updated tests to ensure proper error handling for unsupported MIME types.
* Enhance type safety in OCR file handling
Added type casting for the uploaded file in the _parse_multipart_form function to ensure proper handling of UploadFile instances. This change improves type safety and reduces potential runtime errors during file processing.
* Refactor MIME type handling in document uploads
Updated the MIME type extraction logic to strip parameters from the Content-Type header, ensuring only the base type is used. Added tests to verify that MIME parameters are correctly handled and stripped in various scenarios.
* Update OCR documentation for MIME type recommendations and remove unnecessary tips
Clarified the recommended usage of MIME types for raw bytes in document uploads. Simplified the documentation by removing the tip about multipart file uploads from tools like Postman, ensuring a more concise and focused guide.
* Enhance multipart form handling in OCR endpoints
Updated the _parse_multipart_form function to ignore both 'file' and 'document' fields during form parsing, ensuring that the document built from the uploaded file is not overridden. Added a new test to verify that injected document fields do not affect the constructed document, improving security and robustness of the file upload process.
The /v1/messages/count_tokens proxy endpoint was only passing `messages`
to provider token counting APIs, discarding `system` and `tools`. This
caused clients like Claude Code to receive artificially low token counts
(e.g. 10 instead of 531), preventing proper context window management
and leading to context overflow errors.
Pass system and tools through the full chain:
- TokenCountRequest → proxy_server → provider counters → API handlers
- Bedrock: transform tools to toolConfig format, system to text blocks
- Anthropic/Azure AI: pass through directly (same API format)
Fixes#22285 — extra_headers passed to litellm.image_generation() were
silently dropped on the openai/litellm_proxy/openai_compatible_providers
code path. The azure and azure_ai paths already forwarded them correctly.
