Metadata-Version: 2.1
Name: comfyui-qwen-omni
Version: 0.1.2
Summary: ComfyUI-Qwen-Omni is the first ComfyUI plugin supporting end-to-end multimodal interaction. 
Integrating the Qwen2.5-Omni large multimodal model, it enables seamless joint generation and editing of text, images, audio, and video within the ComfyUI framework. 

Key features:
- Dual-Mode Omni: Supports Qwen2.5-Omni-3B and Qwen2.5-Omni-7B models
- Multi-modal input support: text prompts, images, audio files, and video frames
- Unified output generation: coherent text descriptions and high-quality speech synthesis
- Parameterized control: adjustable temperature, max tokens, sampling strategy
- GPU optimization: 4-bit/8-bit quantization for low-memory environments
- Cross-modal editing: modify content across different media types

By processing multiple input modalities simultaneously, the plugin delivers unprecedented in AI content creation, from creative writing to voiceovers and visual editing. Ideal for artists, developers, and researchers exploring the frontiers of multimodal AI.

Home-page: https://github.com/SXQBW/ComfyUI-Qwen-Omni
Requires-Python: >=3.10
Requires-Dist: accelerate
Requires-Dist: qwen_omni_utils
Requires-Dist: numpy
Requires-Dist: soundfile
Requires-Dist: triton-windows
Requires-Dist: modelscope
Requires-Dist: bitsandbytes
Requires-Dist: pillow
Requires-Dist: requests
