Repackaged model files for ComfyUI. - https://huggingface.co/Qwen/Qwen-Image - https://huggingface.co/Qwen/Qwen-Image-2512 - https://huggingface.co/DiffSynth-Studio/Qwen-Image-Distill-Full Place the files in the following folders
SAVRN's Notes on Qwen-Image_ComfyUI
Plan your storage before your GPUs on this one: the 16 files come to 279 GB on disk, and our record carries no runs_on figure, no parameter count and no per-precision memory number, so we cannot quote a cheapest setup. Comfy Org publishes a repackaging of the Qwen-Image weights into the folders ComfyUI expects, drawing on Qwen-Image, Qwen-Image-2512 and DiffSynth-Studio's Qwen-Image-Distill-Full. Size the card from the files your workflow actually loads, not the full 279 GB.
The repackage carries Apache 2.0, so commercial use, modification and redistribution are permitted with the license and NOTICE files kept and significant changes stated. Our records show it derived from Qwen-Image-Distill-Full, so trace each file to its upstream source and confirm the terms before production. Watch the cadence: released August 5, 2025, last updated August 17, 2026, so pin the file versions you tested rather than pulling fresh.
Model Card
By Comfy Org, published under apache-2.0, revision 7beb7b647f04.
Repackaged model files for ComfyUI. - https://huggingface.co/Qwen/Qwen-Image - https://huggingface.co/Qwen/Qwen-Image-2512 - https://huggingface.co/DiffSynth-Studio/Qwen-Image-Distill-Full Place the files in the following folders
Read Comfy Org's full model card
Qwen-Image
Repackaged model files for ComfyUI.
Original model repository:
- https://huggingface.co/Qwen/Qwen-Image
- https://huggingface.co/Qwen/Qwen-Image-2512
- https://huggingface.co/DiffSynth-Studio/Qwen-Image-Distill-Full
Place the files in the following folders:
ComfyUI/
├── models/
│ ├── diffusion_models/
│ │ ├── qwen_image_distill_full_bf16.safetensors
│ │ ├── qwen_image_distill_full_fp8_e4m3fn.safetensors
│ │ ├── qwen_image_2512_bf16.safetensors
│ │ ├── qwen_image_2512_fp8_e4m3fn.safetensors
│ │ ├── qwen_image_bf16.safetensors
│ │ ├── qwen_image_fp8_e4m3fn.safetensors
│ │ ├── qwen_image_fp8_hq.safetensors
│ │ ├── qwen_image_fp8mixed.safetensors
│ │ └── qwen_image_nvfp4.safetensors
│ ├── text_encoders/
│ │ ├── qwen_2.5_vl_7b.safetensors
│ │ ├── qwen_2.5_vl_7b_fp8_scaled.safetensors
│ │ └── qwen_2.5_vl_7b_nvfp4.safetensors
│ ├── vae/
│ │ └── qwen_image_vae.safetensors
See: https://comfyanonymous.github.io/ComfyUI_examples/qwen_image/
Identity and Version
- Repository
- Comfy-Org/Qwen-Image_ComfyUI
- Publisher
- Comfy Org
- Task
- Not stated by the source
- Modality
- Other
- Library
- diffusion-single-file
- Parameters
- Not stated by the source
- Languages
- Not stated by the source
- Revision
- 7beb7b647f04469fbe64ba8adc2bb0d7e5e9f73f
- First published
- 2025-08-05
- Last updated
- 2026-08-17
Files and Weights
16 files, 279.3 GB in total. The weights are 13 files totalling 279.3 GB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| non_official/diffusion_models/qwen_image_distill_full_bf16.safetensors | Weights | 40.9 GB | 92e35de44229 |
| non_official/diffusion_models/qwen_image_distill_full_fp8_e4m3fn.safetensors | Weights | 20.4 GB | a1d50aa60140 |
| split_files/diffusion_models/qwen_image_2512_bf16.safetensors | Weights | 40.9 GB | cbf55390fff2 |
| split_files/diffusion_models/qwen_image_2512_fp8_e4m3fn.safetensors | Weights | 20.4 GB | 5dc80554d5d8 |
| split_files/diffusion_models/qwen_image_bf16.safetensors | Weights | 40.9 GB | d08fb5d68026 |
| split_files/diffusion_models/qwen_image_fp8_e4m3fn.safetensors | Weights | 20.4 GB | 98763a127701 |
| split_files/diffusion_models/qwen_image_fp8_hq.safetensors | Weights | 22.7 GB | d6027c88c850 |
| split_files/diffusion_models/qwen_image_fp8mixed.safetensors | Weights | 20.5 GB | 0ee8375df948 |
| split_files/diffusion_models/qwen_image_nvfp4.safetensors | Weights | 19.8 GB | 54df91fa379a |
| split_files/text_encoders/qwen_2.5_vl_7b.safetensors | Weights | 16.6 GB | cfafd739459b |
| split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors | Weights | 9.4 GB | cb5636d852a0 |
| split_files/text_encoders/qwen_2.5_vl_7b_nvfp4.safetensors | Weights | 6.1 GB | c7bb24d331f5 |
| split_files/vae/qwen_image_vae.safetensors | Weights | 253.8 MB | a70580f0213e |
| README.md | Documentation | 1.4 KB | — |
| non_official/diffusion_models/README.md | Documentation | 103 B | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 279.3 GB
Released by Comfy Org through its official repository on Hugging Face. Read the license.
Built From
- Derived from DiffSynth-Studio/Qwen-Image-Distill-Full
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 279.3 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About Qwen-Image_ComfyUI
Can I use Qwen-Image_ComfyUI commercially?
Yes. Qwen-Image_ComfyUI is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.