Quantized GGUF versions of the Z-Image Turbo by Tongyi-Mai.
Model Information
Check out the original model card Z-Image Turbo for detailed information about the model.
Usage
The model can be used with:
ComfyUI-GGUF by city96
Description
For that less than 8GB VRAM usage. Needs the GGUF Nodes.
FAQ
Comments (7)
I think you should post Flux.1 AE
Done :)
Hello! Thank you for your work. Do you have a workflow for the GGUF version?
I think a simple workflow should be within the showcase images if you drag them into ComfyUI
I wish I had discovered it sooner!
If you have 6GB VRAM card, the FP8 models work and you don't need to use GGUF. You might need 16GB system RAM and use ComfyUI and the FP8 version of Qwen text encoder, though.
I'm on Neo. Where does this go and what does it do?