uqer1244/MLX-z-image-Base

https://github.com/uqer1244/MLX_z-image-base

This is a 4-bit quantized MLX version of Tongyi-MAI/Z-Image. It is optimized for Apple Silicon (macOS) using the MLX framework.

Model Details

• Transformer: MLX 4-bit quantized
• Text Encoder: MLX 4-bit quantized (Qwen3)
• VAE: Original PyTorch Model (Sourced from original repo)
• Tokenizer: Original Tokenizer (Sourced from original repo)
• Scheduler: MLXFlowMatchEulerScheduler

Usage

This model can be used with the custom MLX pipeline script. Please refer to the original repository for detailed usage instructions regarding the model architecture.

Attribution & License

This model is a derivative work of Tongyi-MAI/Z-Image.

• Original License: Apache 2.0
• Modifications: Converted Transformer and Text Encoder weights to MLX format and quantized to 4-bit.