Z-Image GitHub Repository

Z-Image open source main repository, containing complete model code and documentation, 6B parameter efficient image generation model

Open Source

Code

Documentation

6B Parameters

Visit Resource

Overview

Z-Image is a powerful and efficient image generation model with 6B parameters. Currently there are three variants: Z-Image-Turbo (distilled version, only 8-step inference), Z-Image-Base (base model) and Z-Image-Edit (image editing variant).

Features

6B parameter lightweight model with sub-second inference latency on enterprise-grade H800 GPUs
Runs on consumer devices with 16G VRAM
Photorealistic image generation
Accurate bilingual text rendering in English and Chinese
Strong instruction following capabilities
Scalable Single-Stream DiT (S3-DiT) architecture

Images

Z-Image Elo rating performance on AI Arena

Installation

pip install -e .

Usage

python inference.py

Requirements

Python 3.8+
CUDA 11.0+
8GB+ VRAM (16GB recommended)
Latest diffusers: pip install git+https://github.com/huggingface/diffusers