Modelpublic

Abiray/Wan2.2-LightX2V-260412-4STEP-GGUF

sourceHugging Faceapache-2.0updated 6mo agoView on Hugging Face

6likes2.6kdownloads

Model Card

🎬 Wan2.2 14B I2V Distill LightX2V 4-Step (GGUF)

This repository contains GGUF-quantized versions of the LightX2V Distilled Wan2.2 14B Image-to-Video Model.

By quantizing the ultra-fast 4-step distilled models into the GGUF format, this repository allows for drastically reduced VRAM/RAM requirements while maintaining near real-time, high-performance video generation.

🌟 What's Special About This Version?

—⚡ Ultra-Fast 4-Step Generation: Distillation-accelerated version of Wan2.2 requires only 4 steps instead of the traditional 50+ steps.
—💾 VRAM & Memory Efficient: Utilizing GGUF quantizations (from Q3 up to Q8), you can run full 14B parameter models on consumer-grade hardware.

📦 Available GGUF Models

We provide quantizations ranging from heavily compressed (Q3) to near-lossless (Q8) to fit your specific memory and quality requirements. Both High Noise and Low Noise versions are available.

🎨 High Noise Models (`/high_noise_260412/`)

Filename	Size	Quality Level
`wan2.2_i2v_A14b_high_noise_lightx2v_4step_720p_260412-Q3_K_M.gguf`	7.19 GB	High Compression, Lowest VRAM
`wan2.2_i2v_A14b_high_noise_lightx2v_4step_720p_260412-Q4_K_M.gguf`	9.66 GB	Good Balance
`wan2.2_i2v_A14b_high_noise_lightx2v_4step_720p_260412-Q5_K_M.gguf`	10.8 GB	High Quality, Moderate VRAM
`wan2.2_i2v_A14b_high_noise_lightx2v_4step_720p_260412-Q6_K.gguf`	12.0 GB	Very High Quality
`wan2.2_i2v_A14b_high_noise_lightx2v_4step_720p_260412-Q8_0.gguf`	15.4 GB	Near-Lossless (Recommended if VRAM permits)

🎯 Low Noise Models (`/low_noise_260412/`)

Filename	Size	Quality Level
`wan2.2_i2v_A14b_low_noise_lightx2v_4step_720p_260412-Q3_K_M.gguf`	7.19 GB	High Compression, Lowest VRAM
`wan2.2_i2v_A14b_low_noise_lightx2v_4step_720p_260412-Q4_K_M.gguf`	9.66 GB	Good Balance
`wan2.2_i2v_A14b_low_noise_lightx2v_4step_720p_260412-Q5_K_M.gguf`	10.8 GB	High Quality, Moderate VRAM
`wan2.2_i2v_A14b_low_noise_lightx2v_4step_720p_260412-Q6_K.gguf`	12.0 GB	Very High Quality
`wan2.2_i2v_A14b_low_noise_lightx2v_4step_720p_260412-Q8_0.gguf`	15.4 GB	Near-Lossless (Recommended if VRAM permits)

🎬 Wan2.2 14B I2V Distill LightX2V 4-Step (GGUF)

🌟 What's Special About This Version?

📦 Available GGUF Models

🎨 High Noise Models (/high_noise_260412/)

🎯 Low Noise Models (/low_noise_260412/)

🎨 High Noise Models (`/high_noise_260412/`)

🎯 Low Noise Models (`/low_noise_260412/`)