toxicdog/Qwen-Image-2.1

Qwen-Image 2.1

Repackaged model files for ComfyUI, with INT4 ConvRot converted for testing in Radiant Canvas for fast macOS inference (https://apps.apple.com/us/app/radiant-canvas-ai-image-gen/id6802973075)

Original model repository: https://huggingface.co/Qwen/Qwen-Image-2.1

Place the files in the following folders:

📂 ComfyUI/
└── 📂 models/
    ├── 📂 diffusion_models/
    │   ├── qwen_image_2.1_bf16.safetensors
    │   ├── qwen_image_2.1_int4_convrot.safetensors
    │   └── qwen_image_2.1_int8_convrot.safetensors
    ├── 📂 text_encoders/
    │   ├── qwen3vl_8b_bf16.safetensors
    │   ├── qwen3vl_8b_int4_convrot.safetensors
    │   ├── qwen3vl_8b_int8_convrot.safetensors
    │   └── qwen3vl_8b_w4a8.safetensors
    └── 📂 vae/
        └── qwen_image_2.1_vae_bf16.safetensors

The INT4 ConvRot files use ComfyUI's native ConvRot W4A4 format. Use a recent ComfyUI build and load them with the standard diffusion model and text encoder loaders.