* Fix issue #13995: Handle None metadata in batch requests
- Added null check in add_key_level_controls method to prevent NoneType error
- Updated type hint to Optional[dict] for better type safety
- Added comprehensive test suite to verify the fix works correctly
- All existing tests pass, confirming no regression
Fixes#13995
Co-authored-by: openhands <openhands@all-hands.dev>
* Move test file to tests/test_litellm/proxy/ directory
- Moved test_batch_metadata_none_fix.py from tests/ to tests/test_litellm/proxy/
- Updated import structure to match existing test patterns
- This ensures the test runs in GitHub Actions as requested by @krrishdholakia
Co-authored-by: openhands <openhands@all-hands.dev>
---------
Co-authored-by: openhands <openhands@all-hands.dev>
* fix: pass extra_headers parameter through responses API transformation chain
Ensure extra_headers parameter is properly forwarded from the responses() function
through the transformation handler and config to maintain header propagation in
litellm_completion_request dict.
* Add tests
- Updated the "supports_tool_choice" field to true in the model_prices_and_context_window.json file, allowing for tool choice functionality in the specified model.
Signed-off-by: Ifta Khairul Alam Adil <ikaadil007@gmail.com>
- Cleaned up the code by removing an extra blank line in the GoogleImageGenConfig class.
- This minor adjustment improves code readability without affecting functionality.
Signed-off-by: Ifta Khairul Alam Adil <ikaadil007@gmail.com>
- Remove unnecessary model name prefix stripping
- Directly use the model name in the API URL construction
This change streamlines the URL generation process for the Google AI API, ensuring compatibility with model names without the 'gemini/' prefix.
Signed-off-by: Ifta Khairul Alam Adil <ikaadil007@gmail.com>
- Fix URL construction in Gemini image generation to strip 'gemini/' prefix
- Google AI API expects base model name without the prefix
- Update model references and pricing information for consistency
- Remove outdated image generation pricing entries
Fixes issue where models like 'gemini/imagen-4.0-fast-generate-preview-06-06'
were being rejected by the Google AI API due to incorrect URL formatting.
The optimization eliminates unnecessary conditional branching by replacing the explicit `if-else` structure with a direct return of the boolean expression. Instead of evaluating the condition and then branching to return `True` or `False`, the optimized version directly returns the result of the boolean expression `verbose_logger.isEnabledFor(logging.DEBUG) or set_verbose is True`.
This change removes Python's conditional jump overhead and reduces the number of executed bytecode instructions per function call. The line profiler shows the original version required 3 lines of execution (condition check, conditional return True, fallback return False) while the optimized version executes only 1 line.
The 45% speedup is achieved by:
- **Eliminating branching overhead**: No conditional jumps needed
- **Reducing bytecode instructions**: From ~3 instructions to 1 instruction per call
- **Leveraging Python's short-circuit evaluation**: The `or` operator still evaluates left-to-right and stops early when the first condition is True
The optimization is particularly effective for this logging utility function which is likely called frequently throughout the application. All test cases show consistent 40-75% improvements across different scenarios (debug on/off, verbose flag variations, edge cases with different logging levels), demonstrating the optimization works well regardless of the boolean expression's outcome.
Co-authored-by: codeflash-ai[bot] <148906541+codeflash-ai[bot]@users.noreply.github.com>
