RunDiffusion/Juggernaut-Z-Image
1325.9k
1---2license: cc-by-nc-4.03language:4- en5pipeline_tag: text-to-image6base_model: Tongyi-MAI/Z-Image7tags:8- gguf9- safetensors10- text-to-image11- rundiffusion12- z-image13---14 15<div align="center">16 17<a href="https://www.rundiffusion.com/?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_z_v1&utm_content=header_logo">18 <img src="https://huggingface.co/RunDiffusion/Juggernaut-Z-Image/resolve/main/assets/RD_Mark.png" alt="RunDiffusion" width="110" />19</a>20 21<h1>Juggernaut Z by RunDiffusion</h1>22 23<p><i>A cinematic fine-tune of Z-Image Base — tuned for presentation-ready output.</i></p>24 25<p>26 <a href="https://www.rundiffusion.com/juggernaut-z?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_z_v1&utm_content=cta_primary"><img alt="Try Juggernaut Z" src="https://img.shields.io/badge/%E2%96%B6%20Try%20Juggernaut%20Z-7C3AED?style=for-the-badge&labelColor=7C3AED"></a> <a href="https://www.rundiffusion.com/juggernaut-z-prompt-guide?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_z_v1&utm_content=prompt_guide_badge"><img alt="Prompt Guide" src="https://img.shields.io/badge/Prompt%20Guide-1f1f23?style=for-the-badge"></a> <a href="https://huggingface.co/Tongyi-MAI/Z-Image"><img alt="Base Model: Z-Image" src="https://img.shields.io/badge/%F0%9F%A4%97%20Base%20Model-Z--Image-FFD21E?style=for-the-badge&labelColor=1f1f23"></a> <img alt="License: CC BY-NC 4.0" src="https://img.shields.io/badge/License-CC%20BY--NC%204.0-2ea44f?style=for-the-badge">27</p>28 29</div>30 31<p align="center">32 <img src="https://www.rundiffusion.com/images/juggernaut-z/hero-image.jpg" alt="Juggernaut Z hero" />33</p>34 35> Juggernaut Z is a fine-tune of **Z-Image Base** by **Team Juggernaut**, trained by **KandooAI**, and released through **RunDiffusion**. It is tuned for stronger lighting, sharper focus, more refined skin texture, and more cinematic atmosphere — out of the box.36 37This repository hosts the official RunDiffusion release artifacts: full-precision weights, FP16 and FP8 variants, and a full set of GGUF quantizations.38 39---40 41## Highlights42 43- More dramatic, cinematic **lighting** out of the box44- Sharper **focus** and a more deliberate camera feel45- Cleaner **portraits** with more natural skin texture46- Improved **anatomy** and structural integrity47- Better representation across **ethnicities** by default48- Tuned for editorial, concept, and cinematic work49 50## Comparisons51 52All sets below show **Juggernaut Z (left)** vs **Z-Image Base (right)**. Source: the [RunDiffusion Juggernaut Z announcement](https://www.rundiffusion.com/juggernaut-z?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_z_v1&utm_content=comparison_source).53 54### Lighting55 56More dramatic, cinematic lighting out of the box.57 58596061626364 65### Skin & Texture66 67Cleaner, more natural-looking skin — especially in close-up portraits.68 6970717273 74### Anatomy75 76Cleaner anatomy and more consistent structural detail across a wide range of subjects.77 7879808182 83### Composition84 85Improved subject and object placement within scenes, with further work planned for v2.86 87888990 91### Diversity92 93More balanced results across ethnic backgrounds, with better representation by default.94 9596979899 100### Architecture101 102Cleaner structural lines and more coherent material rendering.103 104105106 107## Recommended Settings108 109| Parameter | Default | Range |110| --- | --- | --- |111| CFG | `6` | `6 – 9` |112| Steps | `35` | `25 – 45` |113 114## Good Fit For115 116- Portraits with cleaner facial detail and stronger focus117- Cinematic scenes with strong lighting and atmosphere118- Concept development and visual exploration119- Editorial and fashion work that benefits from a polished finish120 121## Files In This Repo122 123| File | Format | Notes |124| --- | --- | --- |125| `Juggernaut_Z_V1_by_RunDiffusion.safetensors` | safetensors (bf16) | Original release weights |126| `Juggernaut_Z_V1_by_RunDiffusion_fp16.safetensors` | safetensors (fp16) | Half-precision |127| `Juggernaut_Z_V1_FP8_e4m3fn.safetensors` | safetensors (fp8 e4m3fn) | Lower VRAM footprint |128| `Juggernaut_Z_V1_by_RunDiffusion_q8_0.gguf` | GGUF · q8_0 | Highest-quality quant |129| `Juggernaut_Z_V1_by_RunDiffusion_q6_k-004.gguf` | GGUF · q6_k | |130| `Juggernaut_Z_V1_by_RunDiffusion_q5_k_m-003.gguf` | GGUF · q5_k_m | |131| `Juggernaut_Z_V1_by_RunDiffusion_q5_k_s-005.gguf` | GGUF · q5_k_s | |132| `Juggernaut_Z_V1_by_RunDiffusion_q4_k_m-002.gguf` | GGUF · q4_k_m | |133| `Juggernaut_Z_V1_by_RunDiffusion_q4_k_s-001.gguf` | GGUF · q4_k_s | Smallest footprint |134| `model_index.json` + `transformer/`, `text_encoder/`, `tokenizer/`, `vae/`, `scheduler/` | 🤗 Diffusers format | Loaded by `DiffusionPipeline.from_pretrained("RunDiffusion/Juggernaut-Z-Image")` |135 136Use the `.safetensors` variants with the workflow that matches your local inference stack. Use the `.gguf` variants with a GGUF-compatible runtime. Use the Diffusers component layout with the 🤗 Diffusers library — see below.137 138## Use with 🤗 Diffusers139 140The repo includes `model_index.json` and the standard 🤗 Diffusers component directories (`transformer/`, `text_encoder/`, `tokenizer/`, `vae/`, `scheduler/`) at the root, exported as a `ZImagePipeline`. Load it with:141 142```python143from diffusers import DiffusionPipeline144import torch145 146pipe = DiffusionPipeline.from_pretrained(147 "RunDiffusion/Juggernaut-Z-Image",148 torch_dtype=torch.bfloat16,149).to("cuda")150 151image = pipe(152 "a cinematic portrait, dramatic lighting",153 guidance_scale=6.0,154 num_inference_steps=35,155).images[0]156image.save("output.png")157```158 159`from_pretrained` only downloads files declared in `model_index.json`, so it will not pull the standalone `.safetensors` / `.gguf` variants at the repo root. Requires a version of `diffusers` that includes `ZImagePipeline` support (verified against `diffusers` 0.37.1 and 0.38.0). Commercial use of the model and its outputs is restricted under CC BY-NC 4.0 — see [License & Commercial Use](#license--commercial-use) below.160 161## Links162 163- **Run Juggernaut Z on RunDiffusion** → [rundiffusion.com/juggernaut-z](https://www.rundiffusion.com/juggernaut-z?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_z_v1&utm_content=footer_run)164- **Prompt guide** → [Juggernaut Z Prompt Guide](https://www.rundiffusion.com/juggernaut-z-prompt-guide?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_z_v1&utm_content=footer_prompt_guide)165- **Base model** → [Tongyi-MAI/Z-Image](https://huggingface.co/Tongyi-MAI/Z-Image)166 167## Attribution168 169Juggernaut Z is built on Z-Image Base — credit for the upstream base model belongs to the Z-Image team. This fine-tuned release is by **Team Juggernaut**, with training by **KandooAI**, published by **RunDiffusion**.170 171## License & Commercial Use172 173Juggernaut Z is released under **[CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/)**:174 175- **BY** — attribute RunDiffusion / Team Juggernaut / KandooAI when sharing output.176- **NC** — **non-commercial use only**. You may not use the model — or its outputs in a workflow — for commercial purposes without a license.177 178You are free to fine-tune, merge, build LoRAs, and otherwise modify the model for non-commercial purposes.179 180**For commercial licensing**, custom models, business inquiries, or consultation, contact **[juggernaut@rundiffusion.com](mailto:juggernaut@rundiffusion.com)**.181