✨ Reformulated Announcement
🎉 Coming Soon: Z-Image-Base for On-Site Generation!
Z-Image is a powerful and highly efficient image generation model with 6 billion parameters, currently available in three variants:
🚀 Z-Image-Turbo (This Model)
The distilled version of Z-Image
Key Features:
⚡ Sub-second inference latency on enterprise-grade H800 GPUs
💾 Runs smoothly on consumer devices with 16GB VRAM
🎯 Matches or exceeds leading competitors with only 8 NFEs
🖼️ Excels in photorealistic image generation
🔤 Bilingual text rendering (English & Chinese)
✅ Robust instruction adherence
🧱 Z-Image-Base
The non-distilled foundation model
Designed to unlock the full potential for community-driven fine-tuning and custom development.
✍️ Z-Image-Edit
Fine-tuned variant for image editing tasks
Supports creative image-to-image generation with impressive instruction-following capabilities for precise, natural language-based edits.
🔗 Resources:
SourceLinkComfyUI ModelsComfy-Org/z_imageHuggingFace RepoTongyi-MAI/Z-Image

