Juggernaut Z Fast by RunDiffusion
The fast 4-step Juggernaut Z release for cheaper, faster image generation.
Juggernaut Z Fast is the speed-focused release of Juggernaut Z. It keeps the cinematic, polished Juggernaut look while targeting 4-step generation for lower latency and lower compute cost. Think of it as the Fast/Turbo-style companion to the full Juggernaut Z release.
This repository hosts the official RunDiffusion Fast release artifacts: Fast safetensors variants and GGUF quantizations. The Diffusers export will be added at the repository root when available.
Why Fast
- 4-step generation for quick iteration and lower cost
- Strong Juggernaut Z visual identity with cinematic lighting and polished composition
- Better fit for high-volume workflows where the full model's 25-45 step range is too slow
- Useful for previewing, concept passes, product mockups, social content, and fast prompt exploration
- Available in full Fast weights, FP16, FP8, and GGUF quantized variants
Recommended Settings
| Parameter | Recommendation |
|---|---|
| Steps | 4 |
| Workflow | Use a Fast/Turbo-compatible Z-Image workflow |
| CFG | Start with the workflow default; avoid full-model high-step settings |
| Use case | Fast previews, lower-cost production passes, text-layout checks, product and portrait exploration |
The full Juggernaut Z model is still the better choice when you want slower high-step refinement. Use this Fast release when speed and cost matter.
Files In This Repo
| File | Format | Notes |
|---|---|---|
JuggernautZ_Fast_by_RunDiffusion.safetensors |
safetensors | Original Fast release weights |
Juggernaut_Z_V1_Fast_FP16.safetensors |
safetensors fp16 | Half-precision Fast weights |
Juggernaut_Z_V1_Fast_FP8_e4m3fn.safetensors |
safetensors fp8 e4m3fn | Lower VRAM footprint |
Juggernaut_Z_V1_Fast_by_RunDiffusion_q8_0.gguf |
GGUF q8_0 | Highest-quality quant |
Juggernaut_Z_V1_Fast_by_RunDiffusion_q6_k.gguf |
GGUF q6_k | Balanced quality and size |
Juggernaut_Z_V1_Fast_by_RunDiffusion_q5_k_m.gguf |
GGUF q5_k_m | Medium quant |
Juggernaut_Z_V1_Fast_by_RunDiffusion_q5_k_s.gguf |
GGUF q5_k_s | Smaller q5 quant |
Juggernaut_Z_V1_Fast_by_RunDiffusion_q4_k_m.gguf |
GGUF q4_k_m | Compact quant |
Juggernaut_Z_V1_Fast_by_RunDiffusion_q4_k_s.gguf |
GGUF q4_k_s | Smallest footprint |
Example Gallery
The examples below are pulled from the RunDiffusion asset library for the juggernaut z fast release collection and organized by the categories used during curation.
Click any example image to open the larger CDN version.
| Category | Count |
|---|---|
| People | 30 |
| Text generation | 12 |
| Art & concept | 17 |
| Architecture & interiors | 10 |
| Product & branding | 9 |
| Nature & travel | 4 |
| Food & beverage | 3 |
People
Editorial portraits, fashion, lifestyle, sports, and cinematic character studies.
Text Generation
Readable typography examples: signs, headlines, apparel text, chalkboards, captions, and brand copy.
Art & Concept
Fantasy, sci-fi, surreal scenes, illustration tests, creature design, and concept-art style outputs.
Architecture & Interiors
Modern homes, interiors, classical facades, brutalist structures, and cinematic building studies.
Product & Branding
Product mockups and brand/logo tests showing how Fast handles commercial-style compositions and text-bearing objects.
Nature & Travel
Landscape, travel, outdoor, and environmental examples.
![]() 13 - Rainforest Dog Portrait |
![]() 25 - Sunset Field Silhouette |
![]() 32 - Coastal Cliffs Landscape |
![]() 55 - Blue Scarf Flower Field Fashion |
Food & Beverage
Food, drinks, cafe, cocktail, and still-life examples.
Diffusers
A Fast Diffusers export is planned. When added, it will use the same root-level component layout as RunDiffusion/Juggernaut-Z-Image:
model_index.json
transformer/
text_encoder/
tokenizer/
vae/
scheduler/
Until then, use the standalone Fast .safetensors variants with the workflow that matches your local inference stack, or use the .gguf variants with a GGUF-compatible runtime.
Links
- Run Juggernaut Z on RunDiffusion -> rundiffusion.com/juggernaut-z
- Full Juggernaut Z release -> RunDiffusion/Juggernaut-Z-Image
- Base model -> Tongyi-MAI/Z-Image
Attribution
Juggernaut Z Fast is built on the Z-Image family and released by Team Juggernaut, with training by KandooAI, published by RunDiffusion.
License & Commercial Use
Juggernaut Z Fast is released under CC BY-NC 4.0:
- BY: attribute RunDiffusion / Team Juggernaut / KandooAI when sharing output.
- NC: non-commercial use only. You may not use the model, or its outputs in a workflow, for commercial purposes without a license.
You are free to fine-tune, merge, build LoRAs, and otherwise modify the model for non-commercial purposes.
For commercial licensing, custom models, business inquiries, or consultation, contact juggernaut@rundiffusion.com.
- Downloads last month
- 5
4-bit
5-bit
6-bit
8-bit
Model tree for RunDiffusion/Juggernaut-Z-Image-Fast
Base model
Tongyi-MAI/Z-Image




















































































