Krea 2 Turbo · MFlux Q4
This is a Q4 MFlux version of Krea-2-Turbo.
MFlux runs the latest state-of-the-art generative image models locally on your Mac in native MLX. MFlux is open source and free.
[!NOTE] 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 Krea-2-Turbo
| QUANT | GB | Repo |
|---|---|---|
| BF16 | 34.2 | mflux-community/krea-2-turbo-mflux-bf16 |
| Q8 | 22.2 | mflux-community/krea-2-turbo-mflux-q8 |
| Q6 | 19.0 | mflux-community/krea-2-turbo-mflux-q6 |
| Q5 | 17.4 | mflux-community/krea-2-turbo-mflux-q5 |
| Q4 | 15.8 | this repo |
| Q3 | 14.2 | mflux-community/krea-2-turbo-mflux-q3 |
Converted on 2026-07-13.
Standard MFlux quantization uses MLX affine quantization (4-bit) across the text encoder, transformer and VAE. The quantization level is stored in the weights, so no --quantize flag is needed.
Usage
Install MFlux
uv tool install --upgrade mflux
Generate an image using Krea-2-Turbo Q4:
mflux-generate-krea2 \
--model mflux-community/krea-2-turbo-mflux-q4 \
--base-model krea-2 \
--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/krea-2-turbo-mflux-q4 --local-dir ./krea-2-turbo-mflux-q4
then pass --model ./krea-2-turbo-mflux-q4.
License
This repo inherits the license of the original model: krea-2-community-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 and quantization.
Credits
- Original model: krea/Krea-2-Turbo
- MLX implementation: MFlux by Filip Strand and contributors
- Conversion: MFlux-Community