A streamlined and optimized ComfyUI workflow for the new Z-Image models by Comfy-Org. Built for high-speed generation without the clutter.
Highlights:
Turbo & Standard Support: Easily switch between speed and max quality.
Lora Manager Integrated: Visual and easy management of your Lora library.
Optimized Settings: Pre-configured with
RES_Multistepsampler andCFG 1.0.Clean & Fast: Focused on performance and clarity.
Quick Setup:
Checkpoints: Place in
models/diffusion_models/.Clip/Text Encoders: Use correct Qwen-based models in
models/text_encoders/.VAE: Uses
ae.safetensors.
Full Video Tutorial & Walkthrough: 📺 Watch here: https://youtu.be/QBgB1xAvd34?si=B1YjmxqUJjjZmagq
Description
🚀 Update: Z-Image Turbo (Workflow Cleanup)
What's New:
Streamlined Nodes: Cleaned up the layout and optimized the workflow for better performance.
Note on BASE Model: This version is specifically tuned for Turbo. The CFG is locked within the Sampler Sub-Node for speed, which isn't ideal for the Base model.
Coming Soon: The dedicated Z-Image BASE Upgrade will be released in the next few days! It will feature:
Full CFG Control (essential for Base settings)
Negative Prompt Support
Stay tuned for the Base upgrade! ⚡️