You can deploy this model and start earning money today!

dgrauet/
ernie-image-pe-mlx-q4

Name: ernie-image-pe-mlx-q4
Availability: InStock

baidu/ERNIE-Image-Turbo

finetune

apache-2.0

Views

No views yet

dgrauet/ernie-image-pe-mlx-q4

Int4 quantization (group_size 64, block Linear weights and the token embedding table) of dgrauet/ernie-image-pe-mlx, the MLX conversion of baidu/ERNIE-Image-Turbo.

Quantized with mlx-forge (mlx-forge convert ernie-image-pe --quantize --bits 4).

Usage

These weights can be used with ernie-image-mlx:

bash

pip install ernie-image-mlx
ernie-image-mlx generate \
    -p "a black and white chinese farm dog" \
    --pe-repo-id dgrauet/ernie-image-pe-mlx-q4 \
    -o dog.png

Keep quantize_config.json next to the weights (the loader also infers bits/group_size from the weight shapes if it is missing).

Related Projects

mlx-forge (conversion): https://github.com/dgrauet/mlx-forge
mlx-arsenal (reusable MLX ops): https://github.com/dgrauet/mlx-arsenal
mlx-porting (Claude Code skill): https://github.com/dgrauet/claude-skill-mlx-porting
bf16 variant: https://huggingface.co/dgrauet/ernie-image-pe-mlx

Files

generation_config.json (153.00 B)
pe.safetensors (1.80 GB)
pe_config.json (901.00 B)
quantize_config.json (90.00 B)
split_model.json (896.00 B)
tokenizer.json (16.29 MB)
tokenizer_config.json (20.63 KB)