Model reference · open weights
Qwen-Image is an open-weight image model from city96. Qwen-Image-gguf (GGUF) weighs 13.1 GB; the smallest configuration that runs it is RTX 4090 24 GB.
Summary of the city96/Qwen-Image-gguf model card, 2026-10-01
What it is
| Released by | city96 |
|---|---|
| Released | 2025-08-05 |
| VRAM | 13.1 GB for the weights |
What it runs on
| Card | One image | Memory |
|---|---|---|
| RTX 3060 12 GB … RTX 4060 Ti 16 GB | does not fit | |
| RTX 3090 24 GB | fits | 23.4 GB |
| RTX 4090 24 GB | fits | 23.4 GB |
| RTX 5090 32 GB | fits | 31.0 GB |
| L40S 48 GB | fits | 44.0 GB |
| A100 80 GB | fits | 78.2 GB |
| H100 80 GB | fits | 78.1 GB |
| RTX PRO 6000 Blackwell 96 GB | fits | 93.8 GB |
| DGX Spark (GB10) 128 GB unified | fits | 107 GB |
| H200 141 GB | fits | 138 GB |
| B200 180 GB | fits | 176 GB |
From the model card
This is a direct GGUF conversion of Qwen/Qwen-Image.
The model files can be used in ComfyUI with the ComfyUI-GGUF custom node. Place the required model(s) in the following folders:
| Type | Name | Location | Download |
|---|---|---|---|
| Main Model | Qwen-Image | ComfyUI/models/diffusion_models | GGUF (this repo) |
| Text Encoder | Qwen2.5-VL-7B | ComfyUI/models/text_encoders | Safetensors / GGUF |
| VAE | Qwen-Image VAE | ComfyUI/models/vae | Safetensors |
Example outputs - sample size of 1, not strictly representative
[!NOTE] The Q5_K_M, Q4_K_M and most importantly the low bitrate quants (Q3_K_M, Q3_K_S, Q2_K) use a new dynamic logic where the first/last layer is kept in high precision.
For a comparison, see this imgsli page. With this method, even Q2_K remains somewhat usable.
As this is a quantized model not a finetune, all the same restrictions/original license terms still apply.
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.
How it works
Running it yourself
Rent a machine by the hour. ComfyUI is installed on it. Open ComfyUI through the tunnel, then Templates → Qwen-Image: Text to Image. Its model loader takes .safetensors files: replace it with Unet Loader (GGUF) and choose this model's file there.
# on your rented machine (the ssh line is on its page in the console)
# the GGUF loader nodes, once per machine
git clone https://github.com/city96/ComfyUI-GGUF /opt/ComfyUI/custom_nodes/ComfyUI-GGUF
pip install -r /opt/ComfyUI/custom_nodes/ComfyUI-GGUF/requirements.txt
# get REPO FILE FOLDER: one file into /workspace/models/FOLDER, where ComfyUI loads it from
get() { hf download "$1" "$2" --local-dir /workspace/hf-files && mkdir -p "/workspace/models/$3" && mv "/workspace/hf-files/$2" "/workspace/models/$3/$4"; }
# the model (12.2 GB)
get city96/Qwen-Image-gguf qwen-image-Q4_K_M.gguf unet
# what the template also loads “Qwen-Image: Text to Image” (10.6 GB)
get Comfy-Org/Qwen-Image_ComfyUI split_files/vae/qwen_image_vae.safetensors vae
get Comfy-Org/Qwen-Image_ComfyUI split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors text_encoders
get lightx2v/Qwen-Image-Lightning Qwen-Image-Lightning-8steps-V1.0.safetensors loras
start-comfyui# on your computer, in a second terminal: ComfyUI in your browser at http://localhost:8188
# HOST and PORT are your machine's, from its page in the console
ssh -L 8188:localhost:8188 dev@HOST -p PORT