- Removed AsyncOmniLLM from the main entry points as it is not supported yet.
- Updated the constructor of MultimodalOutputProcessor to set a default value for engine_core_output_type.
- Adjusted output type handling in various methods to utilize the new default engine_core_output_type for consistency.
- Introduced a static method `_dtype_to_name` to map PyTorch data types to their string representations.
- Updated the `dtype_str` assignments in the `OmniProcessor` class to utilize the new mapping method for improved readability and maintainability.
- Removed redundant imports of `torch` and `numpy` in `OmniProcessor` class.
- Improved comments in `parse_singleton_prompt_omni` function for clarity on token handling.
- Ensured `current_platform` is imported only once for better code organization.
- Added logging for exceptions during tensor movement to CPU in OmniRequestState and MultimodalOutputProcessor.
- Improved robustness by ensuring the output pipeline continues without crashing on errors.
- Updated comments for clarity on error handling behavior.
- Introduced `OmniModelRunnerOutput` and `OmniRequestOutput` classes for handling multimodal outputs.
- Implemented `OmniEngineCoreRequest` and `OmniInputPreprocessor` to manage prompt embeddings and additional information.
- Created `OmniTokensPrompt`, `OmniTokenInputs`, and `OmniEmbedsPrompt` for enhanced token handling.
- Added parsing and preprocessing functions to support new input structures.
- Removed the obsolete `diffusion_engine.py` file as part of the refactor.
- Introduced OmniRequestState to manage multimodal request states.
- Enhanced MultimodalOutputProcessor to handle various output types including images, text, and latents.
- Implemented methods for accumulating multimodal tensors and processing outputs.
- Updated output handling to ensure compatibility with vLLM's base processor while allowing custom modality handlers.
- Add comprehensive PRD, architecture design, and test design documents
- Implement core modules: OmniLLM, AsyncOmniLLM, stage configurations
- Add DiT scheduler and cache manager for diffusion models
- Implement CLI integration with --omni flag support
- Add API server and plugin system for vLLM integration
- Create comprehensive test suite with fixtures
- Update dependencies to vLLM 0.10.2 and PyTorch 2.8.0
- Add conda environment setup and package installation
- Implement stage-based processing architecture
- Add multimodal output processing capabilities
This commit establishes the foundation for multi-modality models
inference and serving with non-autoregressive structures.
Gao Han