sdxl unet q8 needs about 3.9 GB of memory at its main file (a 3.0 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 1.9 GB to 5.1 GB (F16).
SDXL UNet only, 2.57B parameters, GGUF format
The SDXL UNet model is a diffusion-based text‑to‑image component with 2.57 billion parameters. It is available in GGUF format with several quantizations. The base model can be used standalone or with a refinement model for higher quality outputs.
Good at
- Text‑to‑image generation
- Latent diffusion refinement
- Standalone UNet use
Limits
- No text encoders or VAE included
- Requires external SDXL checkpoint
- No vision support
Best for
- Generating high‑resolution images
- Fine‑tuning diffusion pipelines
- Research on diffusion models
Who it’s for. Developers and researchers building or extending SDXL pipelines. The model is suitable for those who already have the SDXL checkpoint and need the UNet component in GGUF format.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 1.9 GB | 2.7 GB | Fits | Get file → | |
| — | 3.0 GB | 3.9 GB | Fits | Get file → | |
| F16 | 5.1 GB | 6.3 GB | Fits | Get file → |
For 8 GB VRAM, use the 1.92 GB file (est. 2.7 GB memory). For 12 GB VRAM, the 3.04 GB file (est. 3.9 GB memory) is suitable. For 16 GB VRAM or more, the 5.14 GB F16 file (est. 6.3 GB memory) can be used.
What GPU do you need for sdxl unet q8?
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 the main file (3.0 GB file, about 3.9 GB in use).
- 6 GB VRAM: the best quant that fits is the main file (3.0 GB file, about 3.9 GB in use).
- 8 GB VRAM: the best quant that fits is F16 (5.1 GB file, about 6.3 GB in use).
Which sdxl unet q8 quant should you choose?
If memory is tight, the smallest file is the first one in the table at 1.9 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about sdxl unet q8
How much VRAM does sdxl unet q8 need?
About 3.9 GB at its main file with an 8K context window. That is the 3.0 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 sdxl unet q8 on an 8 GB GPU?
Yes, with F16: the 5.1 GB file needs about 6.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of sdxl unet q8 GGUF?
It depends on the quant: file is 1.9 GB, file is 3.0 GB, F16 is 5.1 GB.
Where can I download sdxl unet q8 GGUF?
The files are on Hugging Face in the AEUPH/sdxl-unet-q8-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 sdxl unet q8 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 the SDXL UNet on 8 GB VRAM?
Yes, the 1.92 GB file requires about 2.7 GB memory at 8k context, which fits in 8 GB VRAM.
Does the model include text encoders?
No, text encoders and VAE are not part of the GGUF files and must be supplied separately.
What license applies to this model?
The model is licensed under apache-2.0.