Instructions to use NoCrypt/SomethingV2_2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use NoCrypt/SomethingV2_2 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("NoCrypt/SomethingV2_2", dtype=torch.bfloat16, device_map="cuda") prompt = "masterpiece, masterpiece, masterpiece, best quality, ultra-detailed, 1girl, hatsune miku, blue hair, upper body, looking at viewer, ?, negative space, bioluminescence, bioluminescence, bioluminescence, darkness, wind, butterfly, black background, portrait, ice" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
Update README.md
Browse files
README.md
CHANGED
|
@@ -86,5 +86,33 @@ Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 3514023396, Size: 456x5
|
|
| 86 |
</details>
|
| 87 |
|
| 88 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 89 |
## Why not call it SomethingV4?
|
| 90 |
Since this model was based on SomethingV2 and there's not THAT much of improvements in some condition. Calling it V4 is just not right at the moment 😅
|
|
|
|
| 86 |
</details>
|
| 87 |
|
| 88 |
|
| 89 |
+
## Non-miku examples
|
| 90 |
+
|
| 91 |
+
<img style="display:inline;margin:0;padding:0;" width="49%" src="https://huggingface.co/NoCrypt/SomethingV2_2/resolve/main/images/00021-4018636341-masterpiece%2C%20best%20quality%2C%201girl%2C%20aqua%20eyes%2C%20baseball%20cap%2C%20blonde%20hair%2C%20closed%20mouth%2C%20earrings%2C%20green%20background%2C%20hat%2C%20hoop%20earr.png"/>
|
| 92 |
+
<img style="display:inline;margin:0;padding:0;" width="49%" src="https://huggingface.co/NoCrypt/SomethingV2_2/resolve/main/images/00022-1334620477-masterpiece%2C%20best%20quality%2C%20landscape.png"/>
|
| 93 |
+
|
| 94 |
+
```yaml
|
| 95 |
+
masterpiece, best quality, 1girl, aqua eyes, baseball cap, blonde hair, closed mouth, earrings, green background, hat, hoop earrings, jewelry, looking at viewer, shirt, short hair, simple background, solo, upper body, yellow shirt
|
| 96 |
+
Negative prompt: EasyNegative
|
| 97 |
+
Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 4018636341, Size: 440x592, Model: SomethingV2_2, Denoising strength: 0.53, Clip skip: 2, ENSD: 31337, Hires upscale: 1.65, Hires steps: 13, Hires upscaler: Latent (nearest-exact)
|
| 98 |
+
```
|
| 99 |
+
|
| 100 |
+
```yaml
|
| 101 |
+
masterpiece, best quality, landscape
|
| 102 |
+
Negative prompt: EasyNegative
|
| 103 |
+
Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 1334620477, Size: 440x592, Model: SomethingV2_2, Denoising strength: 0.53, Clip skip: 2, ENSD: 31337, Hires upscale: 1.65, Hires steps: 13, Hires upscaler: Latent (nearest-exact)
|
| 104 |
+
```
|
| 105 |
+
|
| 106 |
+
## Recommended settings
|
| 107 |
+
- VAE: None (Baked in model, [blessed2](https://huggingface.co/NoCrypt/blessed_vae/blob/main/blessed2.vae.pt))
|
| 108 |
+
- Clip Skip: 2
|
| 109 |
+
- Sampler: DPM++ 2M Karras
|
| 110 |
+
- CFG Scale: 7 ± 5
|
| 111 |
+
- Recommended Positive Prompt: masterpiece, best quality, negative space, (bioluminescence:1.2), darkness, dark background
|
| 112 |
+
- Recommended Negative Prompt: [EasyNegative](https://huggingface.co/datasets/gsdf/EasyNegative)
|
| 113 |
+
- For better results, using hires fix is a must.
|
| 114 |
+
- Hires upscaler: Latent (any variant, such as nearest-exact)
|
| 115 |
+
|
| 116 |
+
|
| 117 |
## Why not call it SomethingV4?
|
| 118 |
Since this model was based on SomethingV2 and there's not THAT much of improvements in some condition. Calling it V4 is just not right at the moment 😅
|