Z-Image Real Skin Portraits
Creates portraits with real human skin texture and natural lighting.
AI model workflow
Z-Image ComfyUI Workflows
Generate portraits and visual concepts from text with Z-Image in ComfyUI. Explore Turbo generation, image-to-image retouching, and compatible LoRA or ControlNet workflows for more control. Run them on RunComfy Cloud, download workflow JSON, or deploy supported workflows as APIs.
10 workflows
Featured workflow
Z-Image | Fast Photorealistic Base Model
The workflows
10 Z-Image ComfyUI Workflows
Compare Z-Image ComfyUI examples for prompt generation, portrait edits, and custom styles. Check the selected variant and required references on the detail page before running or downloading. Run it on RunComfy Cloud, download the workflow JSON, or deploy supported workflows as APIs.
Z-Image Turbo Character Headswap
Seamless head swap tool for perfect character customization.
Z-Image Base Image Generation
Super-fast image maker with stunning clarity and total control.
Z-Image LoRA Inference
Apply your AI Toolkit-trained Z-Image LoRA in ComfyUI through a single RCZimage pipeline node for training-matched output.
Z-Image ControlNet
Total control over image poses, edges, and depth layouts.
Z-Image Turbo Image Generation
Generate ultra-clear visuals fast with unmatched real-time detail.
Z-Image Character Image-to-Image
Turns portraits into lifelike, perfectly detailed realistic faces fast.
Z-Image Turbo LoRA Inference
Run your AI Toolkit-trained Z-Image Turbo LoRA in ComfyUI with training-matched defaults using a single RC custom node.
Z-Image Finetuned Models
Create stunning, detailed images across multiple styles and moods easily.
Z-Image De-Turbo LoRA Inference
Run your AI Toolkit-trained Z-Image De-Turbo LoRA in ComfyUI with training-matched behavior using a single RCZimageDeturbo custom node.
10 of 10 workflows shown
About Z-Image
What can Z-Image do in ComfyUI?
Z-Image in ComfyUI supports text-to-image generation for portraits, photoreal scenes, and visual concepts. Turbo examples focus on fast generation, while dedicated portrait-editing, ControlNet, and LoRA workflows provide different ways to guide an image.
A text prompt is enough for a new image. When the composition or subject needs more guidance, choose an example with image references, ControlNet, or a compatible LoRA. These controls come from the selected setup and its supporting files, not from every Z-Image checkpoint.
Image-to-image retouching and head-swap examples add dedicated editing stages. They should not be confused with a general instruction-editing model simply because they use a source image. Open the relevant example to check which variant, inputs, and add-ons produce the result.
References and model resources: Z-Image GitHub, ComfyUI Z-Image Turbo tutorial, ComfyUI Z-Image examples, Explore Z-Image models on Hugging Face, Explore Z-Image models on Civitai.
Capabilities
Explore Z-Image capabilities in ComfyUI
Start with prompt-based generation, then choose an image reference, structural control, or compatible LoRA when the result needs more direction.
Text-to-image generation
Create a new image from a written scene using a Z-Image generation setup.
Image-to-image portrait editing
Rework a source portrait with a dedicated image-to-image retouching pipeline.
Structure-guided generation
Guide pose, edges, or depth with a compatible Z-Image ControlNet setup.
Z-Image models on RunComfy
Create with Z-Image in a prepared cloud workspace
Open a Z-Image example in ComfyUI with its required model and supporting files prepared. Change prompts and source images, or use the controls included in that setup. Download the JSON for local use after checking its model variant and hardware requirements.
Run a Z-Image workflow nowEditable generation controls
Change the prompt, image dimensions, and settings supported by your chosen example.
Compatible style files
Use the exact model variant and compatible LoRAs or control files required by the setup.
Prepared dependencies
Start from the example's configured environment before adjusting its generation or editing controls.
Cloud GPU options
Choose hardware for your model format, image size, and any added editing or enhancement stages.
Applications
What you can make with Z-Image
Explore natural-looking portraits, character customization, visual styles, and concept variations from the examples below.
Photoreal Portraits
Explore skin texture, lighting, and detail with a portrait-focused setup.
Character Customization
Change a character's head using a dedicated reference-based replacement pipeline.
Multi-Style Images
Explore different looks and moods with compatible fine-tuned Z-Image models.
Frequently asked questions
Z-Image in ComfyUI: FAQ
Choose between Turbo and Base, understand reference edits and compatible add-ons, and check local memory and download requirements.
What Z-Image ComfyUI workflow should I start with?
Choose a Z-Image Turbo generation example if you are starting from a prompt and want quick image variations. Use a portrait-editing, structural-control, or LoRA example when you already know which reference or visual constraint the result needs.
Should I use Z-Image Turbo or Base for my images?
Turbo focuses on fast generation; Base is another starting point for generation and compatible fine-tuned or LoRA setups. Choose the variant used by the example you want to run. Editing examples can add image-to-image or replacement stages, so an editable pipeline does not necessarily mean it uses the separate Z-Image Edit model.
Can Z-Image run on low VRAM with GGUF?
Quantized formats such as GGUF can reduce model memory use when the workflow supports them. That does not guarantee a particular setup will fit your GPU: image size, encoders, and additional stages also matter. Check the selected example's format and requirements, or use a suitable cloud GPU.
Can Z-Image generate readable text in images?
Z-Image Turbo supports Chinese and English text rendering, but readable text is not guaranteed in every output. Keep the requested wording clear and check the generated spelling before using it in a finished asset. For text that must be exact, plan to review or add the final typography separately.
Can I use Z-Image with LoRA and ControlNet?
Yes, with compatible add-ons. Use a Z-Image LoRA example for a trained concept or look, or a ControlNet example for pose, edge, or depth guidance. Match the add-on to the actual base model instead of assuming different model families share the same files.
Can I train my own Z-Image LoRA?
Yes. RunComfy Trainer supports Z-Image Turbo, Z-Image Base, Z-Image De-Turbo, and Z-Image L2P. Train against the same base you plan to use for inference.
Where can I download a Z-Image ComfyUI workflow JSON?
Use Download JSON on a Z-Image card or detail page, then open the file in ComfyUI. A local run also requires the matching checkpoint and any additional model files or custom nodes listed for the example. The JSON contains the workflow structure, not those dependencies.
Related resources
More ways to use Z-Image
Train a custom Z-Image LoRA using a supported Turbo, Base, De-Turbo, or L2P base. Match the training base to the workflow where you intend to use it.
Train a Z-Image Turbo LoRA
Train a fast custom character, product, or style LoRA using the supported Turbo training adapter.
Open Z-Image Turbo TrainerTrain a Z-Image Base LoRA
Use the non-distilled Z-Image base for a compatible custom LoRA training workflow.
Open Z-Image Base TrainerTrain a Z-Image De-Turbo LoRA
Train against the supported de-distilled Z-Image variant for matching inference workflows.
Open De-Turbo TrainerTrain a Z-Image L2P LoRA
Choose the L2P pixel-space base when your intended Z-Image workflow uses that architecture.
Open Z-Image L2P Trainer