sd3.5 large needs about 7.0 GB of memory at Q5_0 (a 5.8 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 6 quants, from 4.8 GB (Q4_0) to 16.3 GB (F16).
Stable Diffusion 3.5 Large quantized for efficient inference
sd3.5 large is a 8.15B parameter diffusion model from StabilityAI, quantized into GGUF format for fast inference. It supports 8k context and can be run with various quantizations from Q4_0 to F16.
Good at
- High‑resolution image generation
- Fast inference with low‑memory quants
- Stable Diffusion 3.5 features
Limits
- No vision support in this GGUF version
- Requires separate clip and vae models
- License is other, not open source
Best for
- Creative image generation on modest GPUs
- Rapid prototyping of diffusion pipelines
- Running in ComfyUI with GGUF nodes
Who it’s for. Developers and artists who need a lightweight Stable Diffusion 3.5 model for local inference. The model is suitable for users with limited VRAM who can choose an appropriate quantization.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_0 | 4.8 GB | 5.8 GB | Fits | Get file → | |
| Q4_1 | 5.3 GB | 6.4 GB | Fits | Get file → | |
| Q5_0 | 5.8 GB | 7.0 GB | Fits | Get file → | |
| Q5_1 | 6.3 GB | 7.5 GB | Tight | Get file → | |
| Q8_0 | 8.8 GB | 10.3 GB | Too big | Get file → | |
| F16 | 16.3 GB | 18.7 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_0 (4.77 GB file, ~5.8 GB memory). If 12 GB is available, Q5_0 (5.77 GB file, ~7 GB memory) offers better quality. Larger GPUs can handle Q8_0 or F16 for maximum fidelity.
What GPU do you need for sd3.5 large?
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_0 (4.8 GB file, about 5.8 GB in use).
- 8 GB VRAM: the best quant that fits is Q5_1 (6.3 GB file, about 7.5 GB in use).
- 12 GB VRAM: the best quant that fits is Q8_0 (8.8 GB file, about 10.3 GB in use).
- 16 GB VRAM: the best quant that fits is Q8_0 (8.8 GB file, about 10.3 GB in use).
- 24 GB VRAM: the best quant that fits is F16 (16.3 GB file, about 18.7 GB in use).
Which sd3.5 large quant should you choose?
Q5_0 (5.8 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_0 at 4.8 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
Source
Frequently asked questions about sd3.5 large
How much VRAM does sd3.5 large need?
About 7.0 GB at Q5_0 with an 8K context window. That is the 5.8 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 sd3.5 large on an 8 GB GPU?
Yes, with Q5_1: the 6.3 GB file needs about 7.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of sd3.5 large GGUF?
It depends on the quant: Q4_0 is 4.8 GB, Q4_1 is 5.3 GB, Q5_0 is 5.8 GB, Q5_1 is 6.3 GB, Q8_0 is 8.8 GB, F16 is 16.3 GB.
Where can I download sd3.5 large GGUF?
The files are on Hugging Face in the calcuis/sd3.5-large-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 sd3.5 large 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 sd3.5 large on 8 GB VRAM?
Yes, the Q4_0 quantization requires about 5.8 GB memory at 8k context, fitting within 8 GB VRAM.
Do I need additional models to run this?
Yes, clip_g.safetensors, clip_l.safetensors, t5xxl_fp8_e4m3fn.safetensors, and diffusion_pytorch_model.safetensors must be placed in the appropriate folders.
Is the model open source?
The license is listed as other; it is not explicitly open source.