martin-rizzo-ComfyUI-ZImagePowerNodes/workflows

Z-Image Power Nodes: Example Workflows

Folders

• /BF16: Contains workflows pre-configured for .safetensors BF16 checkpoints.
• /FP8 : Contains workflows pre-configured for .safetensors FP8 checkpoints.
• /GGUF: Contains workflows pre-configured for GGUF format checkpoints.

Workflows

• z-image-turbo__inpainting.json: Main reference workflow for inpainting tasks using power nodes.
• z-image-turbo__t2i_double_trouble.json: Text-to-image workflow combining two different styles.
• z-image-turbo__text2image.json : Main reference text-to-image generation workflow.

Requirements

To utilize any of these workflows, you need to have ComfyUI configured and the "Z-Image Power Nodes" installed. Additionally, ensure that the required checkpoints (in GGUF or Safetensors format) are placed in the appropriate directories within your ComfyUI setup.

If you choose to use GGUF-format checkpoints, it is necessary to install the "ComfyUI-GGUF" nodes as well, since ComfyUI does not natively support GGUF files. You can find information about these nodes at: https://github.com/city96/ComfyUI-GGUF

Below in this readme, you will find detailed instructions on how to install the necessary nodes and checkpoints.

Please note that while my work with "Z-Image Power Nodes" was developed and tested using the recommended checkpoints listed below, it may also work correctly with other "Z-Image Turbo" checkpoints or when applying LoRA. However, if custom modifications are made to the workflows or alternative checkpoint combinations are used, I cannot guarantee 100% functionality in all cases. It becomes your responsibility to determine the suitable configuration for your customized setups.

The node pack includes an "Advanced" version of "Z-Sampler Turbo" that incorporates additional parameters which may help it function better in your custom workflows.

Checkpoint Files

The following list includes the checkpoint files that I found to perform best during development and testing. While the workflows are compatible with any Z-Image Turbo and Qwen3-4B checkpoints, it's important to note that using other finetuned files may require adjustments to the workflows to extract the best possible quality from those alternative models.

GGUF (Q8/Q5)

• "z_image_turbo-Q5_K_S.gguf" |5.19 GB| Download
  Local Directory: ComfyUI/models/diffusion_models/

• "Qwen3-4B-Q8_0.gguf" |4.28 GB| Download
  Local Directory: ComfyUI/models/text_encoders/

• "ae.safetensors" |335 MB| Download
  Local Directory: ComfyUI/models/vae/

Safetensors (FP8)

• "z-image-turbo_fp8_scaled_e4m3fn_KJ.safetensors" |6.16 GB| Download
  Local Directory: ComfyUI/models/diffusion_models/

• "qwen3_4b_fp8_scaled.safetensors" |4.41 GB| Download
  Local Directory: ComfyUI/models/text_encoders/

• "ae.safetensors" |335 MB| Download
  Local Directory: ComfyUI/models/vae/

Safetensors (BF16)

• "z_image_turbo_bf16.safetensors" |12.3 GB| Download
  Local Directory: ComfyUI/models/diffusion_models/

• "qwen_3_4b.safetensors" |8.04 GB| Download
  Local Directory: ComfyUI/models/text_encoders/

• "ae.safetensors" |335 MB| Download
  Local Directory: ComfyUI/models/vae/

Z-Image Power Nodes

Z-Image Power Nodes can be installed via the ComfyUI Manager or downloaded from its respective repository.

Installation via ComfyUI Manager (Recommended):

• Open ComfyUI and click on the "Manager" button to launch the "ComfyUI Manager Menu".
• Within the ComfyUI Manager, locate and click on the "Custom Nodes Manager" button.
• In the search bar, type "Z-Image Power Nodes".
• Select the option from the search results and click the "Install" button.
• Restart ComfyUI to ensure the changes take effect.

Manual Installation:

For manual installation, follow the instructions provided in the GitHub repository of the project:

• https://github.com/martin-rizzo/ComfyUI-ZImagePowerNodes