CyberRealistic v8 Z-Image Turbo — GGUF FULL Bundle
🖼️ Main advantages - WHY specifically v8 ? (vs examples)
Key changes in v8.0 focus on three areas: structural consistency and anatomy (fewer duplicated limbs, greater stability in complex poses), enhanced detail via SAM3 detailers (faces, eyes, and hands—with greater accuracy at unusual angles), and more natural rendering of materials and lighting.
All images generated locally on a GTX 1070 (8 GB VRAM), 1024×1024.
💡 Tip: These PNGs have the ComfyUI workflow and prompt embedded! Don't just copy the image from the browser preview. Go to the Files and versions tab, open the image there, and click "Download" to get the original file with all metadata intact.
The full ComfyUI workflow used for all examples is available in examples/example_workflow.jpg.
💡 Prompting
If you're coming from SDXL or Pony, the biggest change is simple: describe the image instead of building a tag stack. Z-Image Turbo uses a Qwen3-based text encoder and responds very well to normal descriptive language. Short comma-separated clauses are fine, but every part of the prompt should ideally tell the model something visual.
Instead of: woman, realistic, masterpiece, best quality, detailed skin, cinematic, 8k
Try: A woman standing beside an open apartment window on a warm summer evening, photographed with soft natural light falling across her face, loose dark hair, natural skin texture and an out-of-focus city street behind her.
Original checkpoint downloaded from Civitai: CyberRealistic Z-Image Turbo.
Converted for use with ComfyUI and with stable-diffusion.cpp directly.
📦 What's in this repository
| Component | File | Size | Notes |
|---|---|---|---|
| Diffusion model (light) | zit_CyberRealistic_v8-Q3_K.gguf | 4.31 GB | Q3_K quant — for very tight VRAM budgets |
| Diffusion model (balanced) | zit_CyberRealistic_v8-Q4_K.gguf | 5.15 GB | Q4_K quant — recommended for 8 GB VRAM |
| Diffusion model (compact) | zit_CyberRealistic_v8-Q4_K_S.gguf | 4.83 GB | Q4_K_S — slightly smaller than Q4_K, good middle ground |
| Diffusion model (quality) | zit_CyberRealistic_v8-Q5_K.gguf | 5.68 GB | Q5_K quant — better detail, still fits 8 GB |
| Diffusion model (high) | zit_CyberRealistic_v8-Q6_K.gguf | 6.05 GB | Q6_K quant — for 12+ GB cards |
| Diffusion model (max) | zit_CyberRealistic_v8-Q8_0.gguf | 7.34 GB | Q8_0 quant — near-lossless, needs 12+ GB |
| Text encoder (light) | clip/Qwen3-4B-Q3_K_M.gguf | 2.08 GB | If you're very tight on VRAM |
| Text encoder (best) | clip/Qwen3-4B-Q4_K_M.gguf | 2.50 GB | Optimal balance for 8 GB cards |
| Text encoder (quality) | clip/Qwen3-4B-Q6_K.gguf | 3.31 GB | For 12+ GB cards |
| VAE | vae/ae.safetensors | 335 MB | 16-channel Z-Image VAE |
For a GTX 1070 (8 GB), the recommended combo is: diffusion Q4_K + text encoder Q4_K_M + VAE — total ~8 GB, fits with room to spare.
Note on quantization naming: Z-Image (Lumina2 architecture) does not benefit from mixed
_S/_M/_Lquantization recipes —llama-quantizeproduces byte-identical files to the base_Ktypes. This repository therefore publishes only base quants:Q3_K,Q4_K,Q5_K,Q6_K,Q8_0.
Why this file
The original checkpoint is distributed in FP16 (safetensors, ~11.5 GB), which does not fit into 8 GB VRAM. This Q4_K quantization brings the diffusion model down to ~5.2 GB, leaving room for the text encoder and VAE on mid-range GPUs.
What you need besides this file
Z-Image Turbo is a composite model. This GGUF contains only the diffusion model. To generate images you also need:
- Text encoder:
Qwen3-4B-Q4_K_M.gguf(or the officialqwen_3_4b.safetensors— both work with Power Nodes). - VAE:
ae.safetensorsfrom the original Z-Image release (16-channel VAE).
🚀 ComfyUI workflow
The working path is martin-rizzo/ComfyUI-ZImagePowerNodes. It loads Z-Image GGUF directly and applies the correct latent/conditioning shapes.
Required nodes
Install the Power Nodes pack:
cd ComfyUI/custom_nodes
git clone https://github.com/martin-rizzo/ComfyUI-ZImagePowerNodes.git
Recommended sampler settings
steps: 4 (Turbo is distilled; anything above ~8 starts to burn)
cfg: 1.0
sampler: euler
turbo_creativity: no
old_scheduler: no
noise_injection: no
alternative_refiner: no
width/height: 1024 (must be multiples of 32)
Alternative: standard ComfyUI GGUF nodes
If you really want to use the built-in UnetLoader (GGUF) instead of Power Nodes, you need city96/ComfyUI-GGUF on PR #392 or later:
cd ComfyUI/custom_nodes/ComfyUI-GGUF
git fetch origin pull/392/head:pr392
git checkout pr392
That fixes the x_pad_token shape mismatch. In practice, though, Power Nodes is more reliable for Z-Image Turbo, and that's what's been verified here.
🙏 Credits
Original model: cyberrealistic-z-image-turbo-catalyst by Cyberdelia on Civitai.
Base architecture: Z-Image / Z-Image Turbo by Tongyi Lab (Alibaba).
ComfyUI nodes: martin-rizzo/ComfyUI-ZImagePowerNodes.
Conversion toolchain: city96/ComfyUI-GGUF + llama.cpp fork.
Runtime: leejet/stable-diffusion.cpp.
Quantization and packaging: gimly13.
📄 License
All components in this repository are distributed under Apache 2.0, CreativeML OpenRAIL-M :
Original diffusion checkpoint (Catalyst) — CreativeML OpenRAIL-M.
Qwen3-4B text encoder — Apache 2.0 (Alibaba).
Z-Image VAE (ae.safetensors) — Apache 2.0 (Tongyi Lab).
See the Civitai model page for details on the original model.