Z-Anime ComfyUI Workflow (6GB VRAM) – Anime AI Generator + Easy Installer

Z-Anime is a powerful anime-focused fine-tuned model built on the Z-Image base by Sebastian Böhnke, designed specifically for high-quality anime image generation with strong stylistic consistency, vibrant colors, and clean linework. It excels at producing detailed characters, expressive faces, and polished compositions while maintaining fast inference speeds, especially with its optimized FP8 distillation variant.
 
The models come with a default ComfyUI workflow available on the official Hugging Face repository: https://huggingface.co/SeeSee21/Z-Anime/tree/main/workflows. In this post, I’ve included a slightly modified version of that workflow, optimized specifically for low VRAM setups. These adjustments make it possible to run Z-Anime on GPUs with as little as 6 GB VRAM, improving memory efficiency while maintaining strong anime image quality and consistent results.
 
To simplify setup, a one-click installer is included that automatically installs ComfyUI, required custom nodes, and all essential models directly to your system. This removes the usual complexity of manual installation, dependency setup, and model configuration, giving you a fast and beginner-friendly way to start generating anime art right away.
 
Preloaded Models Within the Installer (Low VRAM)
z-anime-distill-8step-fp8.safetensors – Diffusion Model
https://huggingface.co/SeeSee21/Z-Anime/tree/main/diffusion_models
Qwen3-4B-UD-Q5_K_XL.gguf – Clip Model
https://huggingface.co/unsloth/Qwen3-4B-GGUF/tree/main
ae.safetensors – VAE Model
https://huggingface.co/Comfy-Org/z_image_turbo/tree/main/split_files/vae
2xLexicaRRDBNet_Sharp.pth – Upscaler
https://huggingface.co/Thelocallab/2xLexicaRRDBNet_Sharp/blob/main/2xLexicaRRDBNet_Sharp.pth
1x-SwatKatsLite.pth – Upscaler
https://huggingface.co/Thelocallab/1x-SwatKatsLite/tree/main
The installer omits the standard Z-Anime Image checkpoint models (BF16, FP16) to keep the download size smaller, but you can grab them anytime from the diffusion model link above.
 
Custom Nodes (Portable Windows Setup)
Custom Nodes w/ commands needed for Portable Windows package
ComfyUI Manager - https://github.com/ltdrdata/ComfyUI-Manager
Command to install requirements:
.\python_embeded\python.exe -m pip install -r .\ComfyUI\custom_nodes\ComfyUI-Manager\requirements.txt
Use the ComfyUI Manager’s Install missing custom nodes feature to install any additional nodes that may not work initially after installation.
 
Speed & Performance
On mid-range systems, such as a 6 GB VRAM GPU with 16 GB system RAM, you can render a 1024x1536 image in less 2 minutes using the FP8 quantized model. The workflow is optimized for low VRAM image generation, balancing GPU memory usage and compute load to maintain strong image quality, detailed textures, and consistent outputs even on limited hardware.
 
System Requirements
NVIDIA RTX 30XX / 40XX / 50XX GPU (FP16 supported)
CUDA-compatible GPU (minimum 6 GB VRAM)
Windows OS
At least 20 GB free storage
 
What’s Included
Portable ComfyUI Windows installer (pre-configured)
Automated model + custom node setup
Beginner-friendly layout with expandable advanced controls
 
Usage Notes
Enter a detailed text prompt describing your scene
For best results, expand your prompt using an LLM before pasting into ComfyUI
More descriptive prompts lead to better composition, lighting, and detail in generated anime images
Use the model switch nodes to enable different parts of the workflow such as the checkpoint, LoRA models, and upscaling pipeline for higher-quality outputs