Update README.md

README.md

@@ -1,81 +1,16 @@
-# Wan2.2 GGUF Quantized Models
-
-**Quantized GGUF versions of Wan2.2-I2V-A14B model**  
-This repository contains GGUF-quantized versions of **WAN2.2-I2V-A14B**, converted for use with ComfyUI-GGUF.
-
-## Model Files
-
-- `wan2.2_i2v_high_noise_14B_fp16.gguf`: High-noise model (used for initial denoising steps)
-- `wan2.2_i2v_low_noise_14B_fp16.gguf`: Low-noise model (used for detail refinement)
-
-## Requirements
-
-- [ComfyUI](https://github.com/comfyanonymous/ComfyUI)
-- [ComfyUI-GGUF](https://github.com/city96/ComfyUI-GGUF) extension by city96
-
-## Installation
-
-1. Download both GGUF files and place them in `ComfyUI/models/unet/`
-2. Install ComfyUI-GGUF extension
-3. Restart ComfyUI
-
-## Usage
-
-1. Load the workflow file included in this repository (drag and drop into ComfyUI)
-2. The workflow will automatically use:
-   - High-noise model for initial denoising steps
-   - Low-noise model for final detail refinement
-
-## Quantization Details
-
-- Base model: [Wan-AI/Wan2.2-I2V-A14B](https://huggingface.co/Wan-AI/Wan2.2-I2V-A14B)
-- Quantization method: GGUF (converted using llama.cpp)
-- Quantization level: [Specify your quantization level, e.g., Q5_K_M]
-- Original model size: ~27B parameters (14B active per step)
-- GGUF file sizes: 
-  - high: [Specify size] MB
-  - low: [Specify size] MB
-
-## Why GGUF?
-
-GGUF format allows for:
-- Reduced memory requirements
-- Faster inference on consumer GPUs
-- Compatibility with ComfyUI-GGUF extension
-- Efficient CPU offloading options
-
-## Performance Notes
-
-- For best results, use both high and low noise models as intended by the original architecture
-- Generation quality may vary based on quantization level
-- Recommended GPU: RTX 3090 or better for smooth 720P generation
-
-## Original Model Information
-
-Wan2.2 is an advanced video generative model with:
-- **Effective MoE Architecture**: Separates denoising process with specialized experts
-- **Cinematic-level Aesthetics**: Detailed control over lighting, composition, and color
-- **Complex Motion Generation**: Trained on significantly larger dataset (+83.2% videos)
-- **Efficient High-Definition**: Supports 720P@24fps generation on consumer GPUs
-
-## Community Works
-
-If you've used these GGUF models in your projects, please let us know so we can feature them!
-
-## License Agreement
-
-The original Wan2.2 model is licensed under the Apache 2.0 License. These GGUF conversions are provided under the same terms. You are fully accountable for your use of the models, which must comply with the original license restrictions.
-
-## Acknowledgements
-
-- [Wan-AI](https://huggingface.co/Wan-AI) for the original Wan2.2 model
-- [city96](https://github.com/city96) for ComfyUI-GGUF extension
-- [ggerganov](https://github.com/ggerganov) for llama.cpp
-
-## Contact
-
-For issues with these GGUF files, please open an issue on [GitHub](https://github.com/yourusername/wan2.2-gguf/issues)
-
 ---
-
-2-bit | 3-bit | 4-bit | 5-bit | 6-bit | 8-bit
+license: apache-2.0
+tags:
+  - gguf
+  - wan2.2
+  - i2v
+  - video-generation
+  - wan-ai
+  - comfyui
+language:
+  - en
+  - ja
+  - zh
+library_name: comfyui
+pipeline_tag: image-to-video
+base_model: Wan-AI/Wan2.2-I2V-A14B