A drop-in replacement for 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.
When to use this variant
- 🛠️ You're running fp16 inference and the standard v7 produces black or NaN-corrupted outputs.
- 💻 You're memory-constrained and need fp16 for lower VRAM.
- 🔬 You have fp16 LoRAs or pipelines validated against this fix.
If you're running fp32/bf16 or have no fp16 issues, the standard v7 works fine.
The Newer Juggernauts
| Model | Why upgrade |
|---|---|
| Juggernaut XL v9 | Direct successor on SDXL — RunDiffusion Photo v2. 6M+ downloads. |
| Juggernaut X / XI | Ground-up retrains with GPT-4V captions |
| Juggernaut Z | Lumina-Image-2 architecture |
| Juggernaut Pro Flux | FLUX.1 backbone |
Recommended Settings
Same as standard v7:
| Parameter | Value |
|---|---|
| Resolution | 832 × 1216 (portrait) · 1216 × 832 (landscape) |
| Sampler | DPM++ 2M Karras |
| Steps | 30 – 40 |
| CFG scale | 3 – 7 |
| VAE | Fp16-fixed VAE already baked in |
📖 Prompting library — free to read.
Commercial Use
This model may not be deployed behind paid API services without explicit licensing. Contact juggernaut@rundiffusion.com.
Personal/creative use under CreativeML Open RAIL-M.
Credits
Juggernaut XL v7 was created by KandooAI with RunDiffusion. FP16 VAE fix applied for stable half-precision inference.