Text-to-Image
Diffusers
English
StableDiffusionXLPipeline
stable-diffusion
stable-diffusion-xl
sdxl
photorealistic
fp16
vae-fix
juggernaut
rundiffusion
kandooai
Instructions to use RunDiffusion/Juggernaut-XL-v7-fp16-vae-fix with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use RunDiffusion/Juggernaut-XL-v7-fp16-vae-fix with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("RunDiffusion/Juggernaut-XL-v7-fp16-vae-fix", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
Marketing-flow rewrite: clear 'when to use this variant' section, fp16 troubleshooting context, full CTA flow
Browse files
README.md
CHANGED
|
@@ -1,35 +1,112 @@
|
|
| 1 |
---
|
|
|
|
| 2 |
language:
|
| 3 |
- en
|
| 4 |
-
license: creativeml-openrail-m
|
| 5 |
library_name: diffusers
|
| 6 |
-
|
| 7 |
-
- art
|
| 8 |
-
- people
|
| 9 |
-
- diffusion
|
| 10 |
-
- Cinematic
|
| 11 |
-
- Photography
|
| 12 |
-
- Landscape
|
| 13 |
-
- Interior
|
| 14 |
-
- Food
|
| 15 |
-
- Car
|
| 16 |
-
- Wildlife
|
| 17 |
-
- Architecture
|
| 18 |
-
thumbnail: https://imagedelivery.net/siANnpeNAc_S2q1M3-eDrA/dfaf8264-1355-413a-504d-eb792e69da00/padthumb
|
| 19 |
base_model: stabilityai/stable-diffusion-xl-base-1.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
---
|
| 21 |
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
|
| 26 |
-
|
| 27 |
|
| 28 |
-
|
| 29 |
|
| 30 |
-
Juggernaut
|
| 31 |
|
| 32 |
-
|
| 33 |
|
| 34 |
-
|
| 35 |
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
license: creativeml-openrail-m
|
| 3 |
language:
|
| 4 |
- en
|
|
|
|
| 5 |
library_name: diffusers
|
| 6 |
+
pipeline_tag: text-to-image
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
base_model: stabilityai/stable-diffusion-xl-base-1.0
|
| 8 |
+
tags:
|
| 9 |
+
- stable-diffusion
|
| 10 |
+
- stable-diffusion-xl
|
| 11 |
+
- sdxl
|
| 12 |
+
- text-to-image
|
| 13 |
+
- photorealistic
|
| 14 |
+
- fp16
|
| 15 |
+
- vae-fix
|
| 16 |
+
- juggernaut
|
| 17 |
+
- rundiffusion
|
| 18 |
+
- kandooai
|
| 19 |
---
|
| 20 |
|
| 21 |
+
<div align="center">
|
| 22 |
+
|
| 23 |
+
<a href="https://www.rundiffusion.com/juggernaut?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_xl_v7_fp16&utm_content=header_banner">
|
| 24 |
+
<img src="https://huggingface.co/RunDiffusion/Juggernaut-XL-v7-fp16-vae-fix/resolve/main/assets/Juggernaut_Banner.webp" alt="Juggernaut — by RunDiffusion" />
|
| 25 |
+
</a>
|
| 26 |
+
|
| 27 |
+
<h1>Juggernaut XL v7 (FP16 VAE Fix) by RunDiffusion</h1>
|
| 28 |
+
|
| 29 |
+
<p><i>The fp16-safe variant of v7 — same model, fixed VAE for stable half-precision inference.</i></p>
|
| 30 |
+
|
| 31 |
+
<p>
|
| 32 |
+
<a href="https://app.rundiffusion.com/login?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_xl_v7_fp16&utm_content=hero_cta"><img alt="Try Juggernaut on RunDiffusion — Free" src="https://img.shields.io/badge/%E2%96%B6%20Try%20Juggernaut%20on%20RunDiffusion%20%E2%80%94%20Free-7C3AED?style=for-the-badge&labelColor=7C3AED"></a>
|
| 33 |
+
</p>
|
| 34 |
+
|
| 35 |
+
<p>
|
| 36 |
+
<a href="https://www.rundiffusion.com/prompting?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_xl_v7_fp16&utm_content=prompting_resource"><img alt="Prompting Guides" src="https://img.shields.io/badge/%F0%9F%93%96%20Prompting%20Guides-1f1f23?style=for-the-badge"></a> <a href="https://www.rundiffusion.com/juggernaut?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_xl_v7_fp16&utm_content=lineup_compare"><img alt="Compare the Lineup" src="https://img.shields.io/badge/Compare%20the%20Lineup-1971c2?style=for-the-badge&labelColor=1f1f23"></a> <img alt="FP16 VAE Fix" src="https://img.shields.io/badge/Variant-FP16%20VAE%20Fix-EAB308?style=for-the-badge&labelColor=1f1f23"> <img alt="License: CreativeML Open RAIL-M" src="https://img.shields.io/badge/License-OpenRAIL--M-2ea44f?style=for-the-badge">
|
| 37 |
+
</p>
|
| 38 |
+
|
| 39 |
+
</div>
|
| 40 |
+
|
| 41 |
+
> A drop-in replacement for [**Juggernaut XL v7**](https://huggingface.co/RunDiffusion/Juggernaut-XL-v7) with the **SDXL fp16 VAE fix** baked in. Use this if your fp16 inference produced black images, NaN errors, or noise artifacts on the original v7. Output is otherwise identical.
|
| 42 |
+
|
| 43 |
+
## When to use this variant
|
| 44 |
+
|
| 45 |
+
- 🛠️ You're running **fp16 inference** and the standard v7 produces black or NaN-corrupted outputs.
|
| 46 |
+
- 💻 You're memory-constrained and need fp16 for **lower VRAM**.
|
| 47 |
+
- 🔬 You have **fp16 LoRAs or pipelines** validated against this fix.
|
| 48 |
+
|
| 49 |
+
If you're running fp32/bf16 or have no fp16 issues, the standard [**v7**](https://huggingface.co/RunDiffusion/Juggernaut-XL-v7) works fine.
|
| 50 |
+
|
| 51 |
+
---
|
| 52 |
+
|
| 53 |
+
<div align="center">
|
| 54 |
+
|
| 55 |
+
<h3>🚀 Skip the VAE drama. Try Juggernaut on RunDiffusion.</h3>
|
| 56 |
+
|
| 57 |
+
<p>Every model pre-loaded with the right VAE. <b>No installs. No precision gotchas.</b></p>
|
| 58 |
+
|
| 59 |
+
<a href="https://app.rundiffusion.com/login?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_xl_v7_fp16&utm_content=midpage_cta"><img alt="Launch on RunDiffusion" src="https://img.shields.io/badge/%E2%96%B6%20Launch%20on%20RunDiffusion-7C3AED?style=for-the-badge&labelColor=7C3AED"></a>
|
| 60 |
+
|
| 61 |
+
<p><i>Pre-loaded into ComfyUI · Forge · Automatic1111 · Fooocus · InvokeAI · SwarmUI</i></p>
|
| 62 |
+
|
| 63 |
+
</div>
|
| 64 |
+
|
| 65 |
+
---
|
| 66 |
+
|
| 67 |
+
## The Newer Juggernauts
|
| 68 |
+
|
| 69 |
+
| Model | Why upgrade |
|
| 70 |
+
| --- | --- |
|
| 71 |
+
| [**Juggernaut XL v9**](https://huggingface.co/RunDiffusion/Juggernaut-XL-v9) | Direct successor on SDXL — RunDiffusion Photo v2. **6M+ downloads.** |
|
| 72 |
+
| [**Juggernaut X / XI**](https://huggingface.co/RunDiffusion/Juggernaut-XI-v11) | Ground-up retrains with GPT-4V captions |
|
| 73 |
+
| [**Juggernaut Z**](https://huggingface.co/RunDiffusion/Juggernaut-Z-Image) | Lumina-Image-2 architecture |
|
| 74 |
+
| [**Juggernaut Pro Flux**](https://www.rundiffusion.com/juggernaut-pro-flux?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_xl_v7_fp16&utm_content=cross_promo_pro_flux) | FLUX.1 backbone |
|
| 75 |
+
|
| 76 |
+
## Recommended Settings
|
| 77 |
+
|
| 78 |
+
Same as standard v7:
|
| 79 |
+
|
| 80 |
+
| Parameter | Value |
|
| 81 |
+
| --- | --- |
|
| 82 |
+
| Resolution | `832 × 1216` (portrait) · `1216 × 832` (landscape) |
|
| 83 |
+
| Sampler | `DPM++ 2M Karras` |
|
| 84 |
+
| Steps | `30 – 40` |
|
| 85 |
+
| CFG scale | `3 – 7` |
|
| 86 |
+
| VAE | **Fp16-fixed VAE already baked in** |
|
| 87 |
+
|
| 88 |
+
📖 [**Prompting library**](https://www.rundiffusion.com/prompting?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_xl_v7_fp16&utm_content=prompting_inline) — free to read.
|
| 89 |
+
|
| 90 |
+
## Commercial Use
|
| 91 |
+
|
| 92 |
+
This model **may not be deployed behind paid API services** without explicit licensing. Contact **[juggernaut@rundiffusion.com](mailto:juggernaut@rundiffusion.com)**.
|
| 93 |
+
|
| 94 |
+
Personal/creative use under [CreativeML Open RAIL-M](https://huggingface.co/spaces/CompVis/stable-diffusion-license).
|
| 95 |
+
|
| 96 |
+
## Credits
|
| 97 |
+
|
| 98 |
+
Juggernaut XL v7 was created by **[KandooAI](https://twitter.com/Juggernaut_AI)** with **[RunDiffusion](https://www.rundiffusion.com/?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_xl_v7_fp16&utm_content=credits)**. FP16 VAE fix applied for stable half-precision inference.
|
| 99 |
+
|
| 100 |
+
---
|
| 101 |
|
| 102 |
+
<div align="center">
|
| 103 |
|
| 104 |
+
<h3>Ready to generate?</h3>
|
| 105 |
|
| 106 |
+
<p>Every Juggernaut — one click away on RunDiffusion.</p>
|
| 107 |
|
| 108 |
+
<a href="https://app.rundiffusion.com/login?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_xl_v7_fp16&utm_content=footer_cta"><img alt="Try Juggernaut on RunDiffusion — Free" src="https://img.shields.io/badge/%E2%96%B6%20Try%20Juggernaut%20on%20RunDiffusion%20%E2%80%94%20Free-7C3AED?style=for-the-badge&labelColor=7C3AED"></a>
|
| 109 |
|
| 110 |
+
<p><sub><a href="https://www.rundiffusion.com/juggernaut?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_xl_v7_fp16&utm_content=footer_lineup">Compare the full Juggernaut lineup</a> · <a href="https://www.rundiffusion.com/prompting?utm_source=huggingface&utm_medium=model_card&utm_campaign=juggernaut_xl_v7_fp16&utm_content=footer_prompting">Prompting guides</a> · <a href="mailto:juggernaut@rundiffusion.com">Commercial licensing</a></sub></p>
|
| 111 |
|
| 112 |
+
</div>
|