qwen image 2.1 comfy needs about 7.2 GB of memory at Q4_K_M (a 5.6 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 4 quants, from 5.6 GB (Q4_K_M) to 5.9 GB (Q4_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
Qwen Image 2.1 Prompt Enhancer for ComfyUI
Qwen Image 2.1 Prompt Enhancer is a 9‑billion‑parameter model designed for prompt rewriting in ComfyUI. It supports vision tasks and operates with a 262,144‑token context window. The model is distributed as four Q4_K_M GGUF files and requires matching BF16 vision projectors for image editing.
Good at
- Prompt rewriting for text and image tasks
- Vision‑enabled prompt enhancement
- ComfyUI integration
Limits
- Requires matching BF16 vision projector for I2I
- Not a standalone image‑generation model
- License is other, not open source
Best for
- ComfyUI prompt‑rewrite workflows
- Image‑to‑image editing with reference images
- Text‑to‑image prompt refinement
Who it’s for. Users of ComfyUI who need prompt enhancement for text or image generation. The model is suitable for those who already have the BF16 vision projectors and want to improve prompt quality.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 5.6 GB | 7.2 GB | Tight | Get file → | |
| Q4_K_M | 5.6 GB | 7.2 GB | Tight | Get file → | |
| Q4_K_M | 5.9 GB | 7.5 GB | Tight | Get file → | |
| Q4_K_M | 5.9 GB | 7.5 GB | Tight | Get file → |
The four Q4_K_M GGUF files are 5.63 GB and 5.89 GB each. At an 8 k context size they need about 7.2 GB or 7.5 GB of VRAM respectively. Choose the 5.63 GB file for a 8 GB VRAM budget, and the 5.89 GB file if you have 8.5 GB or more. Memory estimates are approximate.
What GPU do you need for qwen image 2.1 comfy?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: the best quant that fits is Q4_K_M (5.9 GB file, about 7.5 GB in use).
Which qwen image 2.1 comfy quant should you choose?
Q4_K_M is the usual starting point for qwen image 2.1 comfy: it balances file size and quality well. Here it is a 5.6 GB file.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about qwen image 2.1 comfy
How much VRAM does qwen image 2.1 comfy need?
About 7.2 GB at Q4_K_M with an 8K context window. That is the 5.6 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run qwen image 2.1 comfy on an 8 GB GPU?
Yes, with Q4_K_M: the 5.9 GB file needs about 7.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of qwen image 2.1 comfy GGUF?
It depends on the quant: Q4_K_M is 5.6 GB, Q4_K_M is 5.6 GB, Q4_K_M is 5.9 GB, Q4_K_M is 5.9 GB.
What context length does qwen image 2.1 comfy support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.
Where can I download qwen image 2.1 comfy GGUF?
The files are on Hugging Face in the t8star/qwen-image-2.1-comfy repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does qwen image 2.1 comfy use?
Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run this model on 8 GB VRAM?
Yes, the 5.63 GB Q4_K_M file requires about 7.2 GB of VRAM at 8 k context, which fits within an 8 GB GPU.
Do I need a separate vision projector?
Yes, for image‑to‑image tasks you must pair the prompt‑rewriter with the matching BF16 mmproj projector. The prompt‑rewriter alone cannot generate images.
Is this model open source?
The model is released under an "other" license. It is not open source and may have usage restrictions.