MOSS Image Base needs about 4.1 GB of memory at Q4_K_M (a 3.2 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 7 quants, from 2.8 GB (Q3_K_XL) to 10.2 GB (BF16).
5.1B text‑to‑image transformer in GGUF format
MOSS Image Base is a 5.08‑billion‑parameter text‑to‑image model. It supports natural‑language prompts, tag sequences, and JSON layouts for controlling text placement and object arrangement. The model is released under the apache‑2.0 license.
Good at
- Natural‑language image generation
- Layout‑to‑image with JSON prompts
- Text placement control
Limits
- No vision support
- Limited context length (8k tokens)
- Requires CUDA GPU for optimal performance
Best for
- Creating custom image layouts
- Generating images from detailed prompts
- Integrating with ComfyUI or Unsloth Studio
Who it’s for. Developers and artists who need a local, open‑source text‑to‑image model. The model is suitable for projects that require fine‑grained control over image composition and text placement.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q3_K_XL | 2.8 GB | 3.7 GB | Fits | Get file → | |
| Q4_K_M | 3.2 GB | 4.1 GB | Fits | Get file → | |
| Q4_K_XL | 3.4 GB | 4.3 GB | Fits | Get file → | |
| Q5_K_XL | 3.9 GB | 4.9 GB | Fits | Get file → | |
| Q6_K_XL | 4.5 GB | 5.6 GB | Fits | Get file → | |
| Q8_0 | 5.5 GB | 6.6 GB | Fits | Get file → | |
| BF16 | 10.2 GB | 11.9 GB | Too big | Get file → |
For 8‑GB VRAM, use Q4_K_M (3.2 GB, 4.1 GB memory). For 12‑GB VRAM, Q5_K_XL (3.95 GB, 4.9 GB memory) offers better quality. For 16‑GB VRAM or more, Q6_K_XL (4.53 GB, 5.6 GB memory) or Q8_0 (5.48 GB, 6.6 GB memory) provide higher fidelity.
What GPU do you need for MOSS Image Base?
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: the best quant that fits is Q3_K_XL (2.8 GB file, about 3.7 GB in use).
- 6 GB VRAM: the best quant that fits is Q6_K_XL (4.5 GB file, about 5.6 GB in use).
- 8 GB VRAM: the best quant that fits is Q8_0 (5.5 GB file, about 6.6 GB in use).
- 12 GB VRAM: the best quant that fits is BF16 (10.2 GB file, about 11.9 GB in use).
Which MOSS Image Base quant should you choose?
Q4_K_M is the usual starting point for MOSS Image Base: it balances file size and quality well. Here it is a 3.2 GB file.
The Q8_0 file is 5.5 GB, about 1.7 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.
If memory is tight, the smallest file is Q3_K_XL at 2.8 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about MOSS Image Base
How much VRAM does MOSS Image Base need?
About 4.1 GB at Q4_K_M with an 8K context window. That is the 3.2 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 MOSS Image Base on an 8 GB GPU?
Yes, with Q8_0: the 5.5 GB file needs about 6.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of MOSS Image Base GGUF?
It depends on the quant: Q3_K_XL is 2.8 GB, Q4_K_M is 3.2 GB, Q4_K_XL is 3.4 GB, Q5_K_XL is 3.9 GB, Q6_K_XL is 4.5 GB, Q8_0 is 5.5 GB and 1 more.
Where can I download MOSS Image Base GGUF?
The files are on Hugging Face in the unsloth/MOSS-Image-Base-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 MOSS Image Base use?
Hugging Face lists the license as apache-2.0. 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 MOSS Image Base on 8 GB VRAM?
Yes, the Q4_K_M quantized file is 3.2 GB and requires approximately 4.1 GB of VRAM at 8k context.
Does the model support vision inputs?
No, MOSS Image Base is a text‑to‑image model and does not accept image inputs.
What license applies to this model?
The model is released under the apache‑2.0 license.