ST

SDXL

StableGGUF by HyperX-SentienceUpdated 2025-06-24
Download on Hugging Face
Parameters2.57B
Context—
Layers—
Architecturesdxl
Smallest quant1.5 GB
Quick answer

SDXL needs about 2.5 GB of memory at Q5_K_S (a 1.8 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 1.5 GB (Q4_K_S) to 2.8 GB (Q8_0).

Stable Diffusion XL quantized for flexible hardware

SDXL is a latent diffusion model for text‑to‑image generation. The GGUF quantized version offers Q4_K_S, Q5_K_S, and Q8 variants. It can be used standalone or with a refinement model for higher resolution output.

Good at

  • High‑resolution image generation
  • Text‑driven image editing
  • Versatile pipeline integration

Limits

  • No vision support in base model
  • Requires separate refinement for best quality
  • License is openrail++

Best for

  • Generating detailed artwork
  • Creating image variations from prompts
  • Deploying on low‑VRAM devices

Who it’s for. Developers and researchers needing a lightweight SDXL inference solution. The model is suitable for local deployment with GGUF‑compatible engines such as ComfyUI or custom scripts.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_S1.5 GB 2.2 GB
Fits Get file →
Q5_K_S1.8 GB 2.5 GB
Fits Get file →
Q8_02.8 GB 3.6 GB
Fits Get file →

For 2GB+ VRAM use Q4_K_S (file size 1.49 GB, memory 2.2 GB at 8k context). For 3GB+ VRAM use Q5_K_S (1.81 GB, 2.5 GB). For 4GB+ VRAM use Q8 (2.76 GB, 3.6 GB). Estimates are approximate.

What GPU do you need for SDXL?

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 Q8_0 (2.8 GB file, about 3.6 GB in use).

Which SDXL quant should you choose?

Q5_K_S (1.8 GB) is the middle option in this repository. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.

If memory is tight, the smallest file is Q4_K_S at 1.5 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturesdxl
Parameters2.57B

Source

Quantized byHyperX-Sentience
FamilyStable
Licenseopenrail++
Downloads (HF)1,860
Last updated2025-06-24

Frequently asked questions about SDXL

How much VRAM does SDXL need?

About 2.5 GB at Q5_K_S with an 8K context window. That is the 1.8 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 on an 8 GB GPU?

Yes, with Q8_0: the 2.8 GB file needs about 3.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of SDXL GGUF?

It depends on the quant: Q4_K_S is 1.5 GB, Q5_K_S is 1.8 GB, Q8_0 is 2.8 GB.

Where can I download SDXL GGUF?

The files are on Hugging Face in the HyperX-Sentience/SDXL-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 use?

Hugging Face lists the license as openrail++. 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 SDXL on 2GB VRAM?

Yes, the Q4_K_S quantization requires 2GB+ VRAM and uses about 2.2 GB at 8k context.

Does the model support image editing?

The base model can be used for image editing, but higher quality results come from the refinement pipeline.

What license applies to this model?

The model is released under the CreativeML Open RAIL++-M license.

Other GGUFs of stable-diffusion-xl-base-1.0

Similar models

Browse all →

More from HyperX-Sentience

All by HyperX-Sentience →

More Stable models

All Stable →

Other tiny models

All tiny models →