You can deploy this model and start earning money today!

dgrauet/
ernie-image-turbo-mlx-q8

Name: ernie-image-turbo-mlx-q8
Availability: InStock

Views

No views yet

dgrauet/ernie-image-turbo-mlx-q8

Int8 quantization (group_size 64, attention and FFN Linear weights only, leaving the small projections and the VAE in bf16) of dgrauet/ernie-image-turbo-mlx, the MLX conversion of baidu/ERNIE-Image-Turbo.

Quantized with mlx-forge (mlx-forge convert ernie-image --variant turbo --quantize --bits 8).

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

bash

pip install ernie-image-mlx
ernie-image-mlx generate \
    -p "一只黑白相间的中华田园犬" \
    --repo-id dgrauet/ernie-image-turbo-mlx-q8 \
    -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).

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-turbo-mlx
q4 variant: https://huggingface.co/dgrauet/ernie-image-turbo-mlx-q4

model_index.json (547.00 B)
pe_tokenizer_tokenizer_config.json (20.63 KB)
quantize_config.json (107.00 B)
scheduler_scheduler_config.json (482.00 B)
special_tokens_map.json (414.00 B)
split_model.json (1.08 KB)
text_encoder.safetensors (3.39 GB)
text_encoder_config.json (1.64 KB)
tokenizer.json (8.83 MB)
tokenizer_tokenizer_config.json (379.00 B)
transformer.safetensors (8.11 GB)
transformer_config.json (422.00 B)
vae.safetensors (160.33 MB)
vae_config.json (836.00 B)