Iris-3B
Collection
MFlux quantized builds of Iris-3B (undistilled base, 100 steps with guidance) • 6 items • Updated
Iris-3B · MFlux BF16
This is a BF16 MFlux version of iris-3b.
MFlux runs the latest state-of-the-art generative image models locally on your Mac in native MLX. MFlux is open source and free.
Community conversion by MFlux-Community -- not an official release from the original model's authors. The original model's license applies unchanged, see License.
MFlux versions of iris-3b
| QUANT | GB | Repo |
|---|---|---|
| BF16 | 14.0 | this repo |
| Q8 | 8.0 | mflux-community/iris-3b-mflux-q8 |
| Q6 | -- | -- |
| Q5 | -- | -- |
| Q4 | 5.1 | mflux-community/iris-3b-mflux-q4 |
| Q3 | -- | -- |
Converted with MFlux 0.24.0 on 2026-10-27.
BF16 is the full-precision MLX build -- no quantization. Use it as the highest-fidelity reference, or pass --quantize to quantize on the fly.
Usage
Install MFlux
uv tool install --upgrade mflux
Generate an image using iris-3b BF16:
mflux-generate-iris \
--model mflux-community/iris-3b-mflux-bf16 \
--base-model iris-3b-base \
--prompt "A puffin standing on a cliff" \
--width 1024 \
--height 1024 \
--seed 42 \
--output image.png
The weights are downloaded from this repo on first use. To keep a local copy instead:
hf download mflux-community/iris-3b-mflux-bf16 --local-dir ./iris-3b-mflux-bf16
then pass --model ./iris-3b-mflux-bf16.
License
This repo inherits the license of the original model: iris-3b-license. See the original model card for the full terms and any usage restrictions. The weights here were modified from the original release only by conversion to MLX.
Credits
- Original model: speridlabs/iris-3b
- MLX implementation: MFlux by Filip Strand and contributors
- Conversion: MFlux-Community
- Downloads last month
- 43
Hardware compatibility
Log In to add your hardware
Quantized
Model tree for mflux-community/iris-3b-mflux-bf16
Base model
speridlabs/iris-3b