Qwen Image 2.1 DiT needs about 5.4 GB of memory at Q4_K_M (a 4.3 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 3 quants, from 4.3 GB (Q4_K_M) to 7.7 GB (Q8_0).
Qwen Image 2.1 DiT – 7B parameter text‑to‑image and image editing model
Qwen Image 2.1 DiT is a 7.12B parameter diffusion transformer for text‑to‑image generation and editing. It supports transparent image generation, up to 10 reference images, and local edits via masks or annotations. The model is available in GGUF format with several quantizations.
Good at
- Text‑to‑image generation
- Transparent image creation
- Image editing with reference images
Limits
- No vision support beyond image generation
- Non‑commercial license only
- Requires Qwen‑Image‑2.1 Text Encoder
Best for
- Generating high‑quality images from text prompts
- Editing existing images with reference guidance
- Creating transparent (RGBA) images
Who it’s for. Developers and researchers building text‑to‑image pipelines who need a lightweight, unified model for generation and editing. The model is suitable for local inference with GGUF support via llama.cpp or ComfyUI.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 4.3 GB | 5.4 GB | Fits | Get file → | |
| Q6_K | 6.0 GB | 7.2 GB | Tight | Get file → | |
| Q8_0 | 7.7 GB | 9.1 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_K_M (file size 4.34 GB, estimated 5.4 GB memory at 8 k context). For 12 GB VRAM, Q6_K (6 GB, 7.2 GB memory) is suitable. If more VRAM is available, Q8_0 (7.69 GB, 9.1 GB memory) provides the highest precision.
What GPU do you need for Qwen Image 2.1 DiT?
These figures add the file size, an fp16 KV cache at an 8K context window 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: the best quant that fits is Q4_K_M (4.3 GB file, about 5.4 GB in use).
- 8 GB VRAM: the best quant that fits is Q6_K (6.0 GB file, about 7.2 GB in use).
- 12 GB VRAM: the best quant that fits is Q8_0 (7.7 GB file, about 9.1 GB in use).
Which Qwen Image 2.1 DiT quant should you choose?
Q4_K_M is the usual starting point for Qwen Image 2.1 DiT: it balances file size and quality well. Here it is a 4.3 GB file.
The Q8_0 file is 7.7 GB, about 1.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Qwen Image 2.1 DiT
How much VRAM does Qwen Image 2.1 DiT need?
About 5.4 GB at Q4_K_M with an 8K context window. That is the 4.3 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 DiT on an 8 GB GPU?
Yes, with Q6_K: the 6.0 GB file needs about 7.2 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 DiT GGUF?
It depends on the quant: Q4_K_M is 4.3 GB, Q6_K is 6.0 GB, Q8_0 is 7.7 GB.
Where can I download Qwen Image 2.1 DiT GGUF?
The files are on Hugging Face in the pottokao/Qwen-Image-2.1-DiT-GGUF 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 DiT 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 Qwen Image 2.1 DiT on 8 GB VRAM?
Yes, the Q4_K_M quantization requires about 5.4 GB of VRAM at 8 k context, which fits within an 8 GB budget.
Does the model support image editing?
Yes, it supports up to 10 reference images and local edits via circles, painted annotations, or masks.
What license applies to this model?
The model is licensed under the Qwen Research License Agreement and is for non‑commercial use only.