Sakura Iris 3B Uncensored needs about 4.3 GB of memory at Q4_K_M (a 3.4 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 6 quants, from 2.5 GB to 4.5 GB (Q8_0).
Sakura Iris 3B Uncensored – 3B pixel‑space diffusion with an uncensored text encoder
Sakura Iris 3B Uncensored is a 2.99 B parameter pixel‑space diffusion transformer. It outputs 1024×1024 images directly, without a VAE, and uses an uncensored Qwen3‑VL‑4B‑Instruct text encoder. The model is released under Apache‑2.0.
Good at
- Pixel‑space image generation
- Dense vision tasks like depth estimation
- Image restoration and upscaling
Limits
- No VAE support – do not pass --vae
- No vision‑only inference – requires text prompt
- Requires 8 k context for full operation
Best for
- Generating high‑resolution images from text
- Fine‑tuning for vision tasks
- Creating detailed concept art
Who it’s for. Developers and researchers who need a lightweight pixel‑space diffusion model for image generation and vision tasks. The model is suitable for those who can manage its memory requirements and are comfortable with an uncensored text encoder.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 2.5 GB | 3.3 GB | Fits | Get file → | |
| Q4_K_M | 3.4 GB | 4.3 GB | Fits | Get file → | |
| Q5_K_M | 3.6 GB | 4.6 GB | Fits | Get file → | |
| Q6_K | 3.9 GB | 4.9 GB | Fits | Get file → | |
| — | 4.3 GB | 5.3 GB | Fits | Get file → | |
| Q8_0 | 4.5 GB | 5.5 GB | Fits | Get file → |
For 8 GB VRAM, use Q4_K_M (3.37 GB file, ~4.3 GB memory at 8k context). For 12 GB VRAM, Q5_K_M (3.65 GB file, ~4.6 GB memory). For 16 GB VRAM, Q6_K (3.93 GB file, ~4.9 GB memory). File sizes are approximate and memory estimates are approximate.
What GPU do you need for Sakura Iris 3B Uncensored?
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 (2.5 GB file, about 3.3 GB in use).
- 6 GB VRAM: the best quant that fits is Q8_0 (4.5 GB file, about 5.5 GB in use).
Which Sakura Iris 3B Uncensored quant should you choose?
Q4_K_M is the usual starting point for Sakura Iris 3B Uncensored: it balances file size and quality well. Here it is a 3.4 GB file.
The Q8_0 file is 4.5 GB, about 1.3 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 the first one in the table at 2.5 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 Sakura Iris 3B Uncensored
How much VRAM does Sakura Iris 3B Uncensored need?
About 4.3 GB at Q4_K_M with an 8K context window. That is the 3.4 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 Sakura Iris 3B Uncensored on an 8 GB GPU?
Yes, with Q8_0: the 4.5 GB file needs about 5.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Sakura Iris 3B Uncensored GGUF?
It depends on the quant: file is 2.5 GB, Q4_K_M is 3.4 GB, Q5_K_M is 3.6 GB, Q6_K is 3.9 GB, file is 4.3 GB, Q8_0 is 4.5 GB.
Where can I download Sakura Iris 3B Uncensored GGUF?
The files are on Hugging Face in the webmp3/Sakura-Iris-3B-Uncensored-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 Sakura Iris 3B Uncensored 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 Sakura Iris 3B Uncensored on 8 GB VRAM?
Yes, with the Q4_K_M quantization the file is 3.37 GB and requires about 4.3 GB of VRAM at 8k context.
Does the model support image‑only inference?
No, it requires a text prompt and an uncensored text encoder; it has no VAE.
What license does the model use?
Apache‑2.0. The model is released under the Apache‑2.0 license.