DFloat11/Wan2.1-T2V-14B-Diffusers-DF11

DFloat11 Compressed Model: Wan-AI/Wan2.1-T2V-14B-Diffusers

This model uses DFloat11 lossless compression. It's 30% smaller than the original BFloat16 model, yet produces bit-identical outputs and runs efficiently on GPUs.

📊 Performance Comparison

Metric	Wan2.1-T2V-14B (BFloat16)	Wan2.1-T2V-14B (DFloat11)
Model Size	28.64 GB	19.39 GB
Peak GPU Memory (2s 480p Video)	30.79 GB	22.22 GB
Generation Time (an A100 GPU)	339 seconds	348 seconds

🔍 How It Works

We apply Huffman coding to the exponent bits of BFloat16 model weights, which are highly compressible. We leverage hardware-aware algorithmic designs to enable highly efficient, on-the-fly weight decompression directly on the GPU. Find out more in our research paper.

🔧 How to Use

A complete usage guide is available in our GitHub repository: https://github.com/LeanModels/DFloat11/tree/master/examples/wan2.1.

📄 Learn More

• Paper: 70% Size, 100% Accuracy: Lossless LLM Compression for Efficient GPU Inference via Dynamic-Length Float
• GitHub: https://github.com/LeanModels/DFloat11
• HuggingFace: https://huggingface.co/DFloat11