ERNIE-Image Turbo · MFlux Q8

This is a Q8 MFlux version of ERNIE-Image-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.

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 ERNIE-Image-Turbo

QUANT	GB	Repo
BF16	23.1	mflux-community/ernie-image-turbo-mflux-bf16
Q8	12.4	this repo
Q6	9.5	mflux-community/ernie-image-turbo-mflux-q6
Q5	8.1	mflux-community/ernie-image-turbo-mflux-q5
Q4	6.6	mflux-community/ernie-image-turbo-mflux-q4
Q3	5.2	mflux-community/ernie-image-turbo-mflux-q3

Converted on 2026-08-04.

Standard MFlux quantization uses MLX affine quantization (8-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 ERNIE-Image-Turbo Q8:

mflux-generate-ernie-image-turbo \
  --model mflux-community/ernie-image-turbo-mflux-q8 \
  --base-model ernie-image-turbo \
  --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/ernie-image-turbo-mflux-q8 --local-dir ./ernie-image-turbo-mflux-q8

then pass --model ./ernie-image-turbo-mflux-q8.

License

This repo inherits the license of the original model: apache-2.0. 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.