CoolFace
Modelpublic

methelina/VBVR-wan2.2-I2V-14B-high-SNR-Calibrated-Hybrid-GGUF

sourceHugging Faceapache-2.0updated 6mo agoView on Hugging Face
2likes98downloads
Model Card

VBVR: A Very Big Video Reasoning Suite – GGUF

GGUF quantized version of LiconStudio/VBVR-wan2.2-comfy-bf16 Optimized for CPU and GPUs with limited VRAM.

Quantization Method

GGUF quantization made by my DIY framework - HQFbyLL (Hot Quantization Framework) :) MB I'll released it someday...

📊 Specification

QuantizationFile Size
Q4KS~8.5 GB
Q5KM~10.5 GB
Q8_0~15.4 GB

Other coming soon...

🛠️ ComfyUI Usage

  1. 1.Install ComfyUI-GGUF or ComfyUI-KJNodes(My Best XD) custom node.
  2. 2.Place .gguf into ComfyUI/models/models/diffusion_models/.
  3. 3.Use UnetLoaderGGUF or Diffusion Model Loader KJ node to load the model.
  4. 4.Profit!

⚖️ License

Wan-AI Software License – see LICENSE.txt