ComfyUI: Local image generation using Z-Image Turbo
ComfyUI
Demo summary
The video shows the process of generating an image locally in ComfyUI, explaining parameters like shift value, step count, and CFG for the Z-Image model.
Step-by-step
- Specify the width and height of your image
- Set the batch size for the number of images to generate at once
- Enter your prompt describing the desired image
- Adjust the shift value to control contrast and detail
- Set the number of steps for generation
- Configure the CFG value to determine prompt adherence
- Select the sampler name and scheduler algorithm
- Click Run to generate the image
Options
- Generate high resolution images up to 4K
- Lower shift value to 1 for higher contrast and detail
- Increase shift value to 6 for softer images with less detail
- Set CFG between 0.5 and 1 for softer, less saturated images
- Increase CFG to 1.x or 2 to improve text accuracy or color saturation
- Use Euler with DDIM uniform for more variation from the same prompt
Watch out for
- Quality starts to degrade if generating above 2048x2048
- Z-Image Turbo is designed for a CFG of 1; going above 3 will mess up the image
- Negative prompts are ignored when using a CFG of 1
Tips
- The sweet spot for step count is 8 to 10 steps
- Use 4 to 6 steps for faster generation, though images will be softer and noisier
- Avoid setting steps significantly above 10 as it can make the image too sharp
- Leave sampler and scheduler at default (res multistep and simple) for generally good results
Highlights
“This was super quick. This only took like 14 seconds on my GPU with 16 GB of VRAM.”
All demos from “The best free AI image generator is here!”
- 22:240:45Generate images with Z-Image Turbo on Hugging FaceThe creator demonstrates how to use the Z-Image Turbo web interface on Hugging Face to generate a realistic photo of a minimalist room by entering a prompt and adjusting resolution settings.Hugging Face· Text to Image
- 24:211:50Setting up Z-Image Turbo in ComfyUIA walkthrough of loading the Z-Image Turbo JSON workflow into ComfyUI, updating the manager, and selecting the required text encoder and VAE models.ComfyUI· AI Image Generator
- 26:113:39Local image generation using Z-Image TurboCurrentThe video shows the process of generating an image locally in ComfyUI, explaining parameters like shift value, step count, and CFG for the Z-Image model.ComfyUI· AI Image Generator
- 30:512:19Running quantized Z-Image GGUF for low VRAMThe creator demonstrates how to use a compressed GGUF version of Z-Image in ComfyUI to generate images on hardware with as little as 4GB of VRAM.ComfyUI· AI Image Generator
- 33:282:46Image-to-image transformation with Z-ImageThe user demonstrates how to modify a ComfyUI workflow to perform image-to-image tasks, converting a 3D animation into a realistic photo using Z-Image Turbo.ComfyUI· Image to Image
- 37:581:16Applying LoRAs to Z-Image workflowsA demonstration of adding a LoRA node to a Z-Image workflow in ComfyUI to generate specific styles, specifically showing a Commodore 64 pixel art output.ComfyUI· AI Image Generator
- Watch “The best free AI image generator is here!” →
AI Image Generator
1:560:39Load a text-to-image workflow and download missing modelsThe user demonstrates how to load a default text-to-image template in ComfyUI and use the 'show missing models' feature to download required AI checkpoints.Kevin Stratvert- 4:110:29Manage generated assets in ComfyUIThe presenter demonstrates how to access the assets panel using the 'A' shortcut to view, save, or delete previously generated images.Kevin Stratvert
15:005:45Generate image with Hermes Agent and ComfyUIThe host uses the Hermes Agent TUI to request a 1024x1024 image of a tree using the Zimage Turbo model, which the agent executes by generating and running a ComfyUI workflow.ComfyUI
6:420:59Configure Gemma 2B Text Encoder in ComfyUIThe creator shows how to load the Gemma 2B text encoder model into the ComfyUI workflow and select it from the node dropdown menu.AI Search
4:270:30Copying and replacing prompts in ComfyUIThe user demonstrates a custom node feature that allows hovering over prompt examples to copy and then replace the current prompt in the workflow with a single click.pixaroma- 13:030:51Managing node colors with PixaRoma nodesThe video demonstrates UI updates for PixaRoma nodes in ComfyUI, including selecting color swatches, copying/pasting colors between nodes, and setting favorite colors.pixaroma
4:150:35Launch ComfyUI on WindowsThe creator demonstrates how to launch the portable version of ComfyUI by running the 'run_nvidia_gpu' batch file and accessing the interface via a web browser.WINBUSH- 10:071:30Browse and load ComfyUI templatesThe creator demonstrates how to navigate the templates menu, filter for local models, and load a Flux 2 image generation workflow.WINBUSH
0:000:33Configuring MSR mode in LTX DirectorThe user demonstrates setting up the Director node in ComfyUI for MSR (Multiple Subject Reference) mode, including bypassing clean latent slices and enabling the MSR LoRA.RedShade- 5:090:19Loading LTX Director V5 workflow templateThe video shows how to search for and load the LTX Director V5 template within ComfyUI and input prompts for character generation.RedShade
4:251:21Iterative rerolling at specific generation stagesThe video shows how to use the 'reroll' button at specific stages of the workflow to change motion or visual quality without re-generating the initial seed samples.Fox•Fur•Essence Films
9:173:04Install ComfyUI using Easy InstallThe instructor demonstrates how to download the ComfyUI Easy Install zip from GitHub, extract it, and run the BAT file to set up a portable local installation.pixaroma