Qwen Image 2.1 PE MTP needs about 7.6 GB of memory at Q4_K_M (a 6.0 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 8 quants, from 6.0 GB (Q4_K_M) to 9.8 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 4.4 GB.
Qwen Image 2.1 PE MTP – prompt enhancer for image generation and editing
Qwen Image 2.1 PE MTP is a 9.2 B‑parameter prompt enhancer built on Qwen‑Image‑2.1‑PE‑I2I. It adds a multi‑token prediction head for speculative decoding in llama.cpp. The model supports vision input and can rewrite vague editing instructions into precise prompts.
Good at
- Vision‑based prompt rewriting
- Supports multi‑image editing
- Compatible with llama.cpp speculative decoding
Limits
- No native vision support in the base GGUF; requires separate vision file
- Requires 8–12 GB VRAM for Q4_K_M, 16 GB for Q8_0
- License is Qwen Research License – not open source
Best for
- Enhancing text prompts for image generation
- Preparing prompts for multi‑image editing
- Using speculative decoding in ComfyUI or llama-server
Who it’s for. Developers and researchers who need a lightweight prompt enhancer for Qwen‑Image‑2.1. The model is useful when working with llama.cpp or ComfyUI and requires a separate vision encoder for image input.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 6.0 GB | 7.6 GB | Tight | Get file → | |
| Q4_K_M | 6.0 GB | 7.6 GB | Tight | Get file → | |
| Q4_K_M | 6.0 GB | 7.6 GB | Tight | Get file → | |
| Q4_K_M | 6.0 GB | 7.6 GB | Tight | Get file → | |
| Q8_0 | 9.8 GB | 11.4 GB | Too big | Get file → | |
| Q8_0 | 9.8 GB | 11.4 GB | Too big | Get file → | |
| Q8_0 | 9.8 GB | 11.4 GB | Too big | Get file → | |
| Q8_0 | 9.8 GB | 11.4 GB | Too big | Get file → |
For 8–12 GB VRAM, use Q4_K_M (≈5.99 GB file, ~7.6 GB memory at 8 k context). For 16 GB VRAM or more, use Q8_0 (≈9.79 GB file, ~11.4 GB memory at 8 k context). Memory estimates are approximate.
What GPU do you need for Qwen Image 2.1 PE MTP?
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 (6.0 GB file, about 7.6 GB in use).
- 12 GB VRAM: the best quant that fits is Q8_0 (9.8 GB file, about 11.4 GB in use).
Which Qwen Image 2.1 PE MTP quant should you choose?
Q4_K_M is the usual starting point for Qwen Image 2.1 PE MTP: it balances file size and quality well. Here it is a 6.0 GB file.
The Q8_0 file is 9.8 GB, about 1.6 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 PE MTP
How much VRAM does Qwen Image 2.1 PE MTP need?
About 7.6 GB at Q4_K_M with an 8K context window. That is the 6.0 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 PE MTP on an 8 GB GPU?
Yes, with Q4_K_M: the 6.0 GB file needs about 7.6 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 PE MTP GGUF?
It depends on the quant: Q4_K_M is 6.0 GB, Q4_K_M is 6.0 GB, Q4_K_M is 6.0 GB, Q4_K_M is 6.0 GB, Q8_0 is 9.8 GB, Q8_0 is 9.8 GB and 2 more.
What context length does Qwen Image 2.1 PE MTP 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 4.4 GB at 32K.
Where can I download Qwen Image 2.1 PE MTP GGUF?
The files are on Hugging Face in the mozophe/Qwen-Image-2.1-PE-MTP-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 PE MTP 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 PE MTP on 8 GB VRAM?
Yes, the Q4_K_M quantization requires about 7.6 GB of VRAM at 8 k context, which fits within 8 GB.
Does this model support vision input directly?
No, it requires a separate vision file from the base Qwen‑Image‑2.1‑PE‑I2I model.
What license applies to this model?
The model is licensed under the Qwen Research License Agreement, not an open source license.