Onyx Z Image Turbo — 4-bit MLX
A 4-bit MLX quantization of the official Tongyi-MAI/Z-Image-Turbo, built for
on-device inference in the Onyx app with a documented, reproducible pipeline.
- Source: official
Tongyi-MAI/Z-Image-Turbo(diffusers), quantized in-house. - Quantization: 4-bit, group 64, MLX. Only transformer + text-encoder Linear leaves are quantized;
norms/embeddings/convs and the whole VAE stay fp16. Components:
transformer/,text_encoder/(Qwen3),vae/(16-ch FLUX-family),tokenizer/,scheduler/. - Inference: ~8 steps, CFG off (distilled).
- Verified: reloaded through the Onyx engine and render-checked before publish.
Quantized derivative for efficient on-device use; all credit for the model to Tongyi-MAI (Alibaba).