uqer1244/MLX_z-image

MLX version of z-image model

44
/ 100
Emerging

Implements Z-Image-Turbo for Apple Silicon with 4-bit quantization and optimizations including fused QKV projections, hardware-accelerated Flash Attention, and cached rotary embeddings to minimize memory bandwidth bottlenecks. The pipeline integrates a native MLX transformer with PyTorch-based text encoder and VAE components, supporting LoRA adapters and ComfyUI nodes for local image generation on Mac devices.

No Package No Dependents
Maintenance 10 / 25
Adoption 8 / 25
Maturity 13 / 25
Community 13 / 25

How are scores calculated?

Stars

46

Forks

6

Language

Python

License

Apache-2.0

Category

flux-model-tools

Last pushed

Feb 21, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/uqer1244/MLX_z-image"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.