---
license: creativeml-openrail-m
---
NAIの含まれる各種モデル(any系)をACertaintyベースで再現しようという試みです。  
レシピに表記されているチェックポイントにNAIが含まれていなければNAIリークフリーのモデルになります。  
マージはよく分からないので適当にマージしてください。  
蒸留画像は使用していません。下記のレポジトリからデータセットのキャプションのみダウンロードできます。  
DataSet: [https://huggingface.co/datasets/paimonimpact/ONN](https://huggingface.co/datasets/paimonimpact/ONN)    
## ONN_anyV3.fp16.safetensors
![](https://huggingface.co/paimonimpact/ONN/resolve/main/images/oon001.jpg)
▼Use Models
- ACertainty.ckpt
- bp_1024_e10.ckpt

ACertaintyにany_A ~ FのデータセットでDB。  

| Model: A     | Model: B | Weight                                                                | Base alpha | Merge Name          |
| ------------ | -------- | --------------------------------------------------------------------- | ---------- | ------------------- |
| A | C | - | 0.3          | AC |
| B | E | - | 0.5          | BE |
| D | F | - | 0.5          | DF |
| BE | DF | 0.0,0.0,0.1,0.3,0.5,0.7,1.0,0.7,0.5,0.3,0.1,0.0,0.0,0.0,0.1,0.3,0.5,0.7,0.5,0.3,0.1,0.0,0.0,0.0,0.0 | 0.3          | BEDF |
| BEDF | AC | 1.0,0.9,0.7,0.5,0.3,0.1,0.3,0.3,0.3,0.3,0.3,0.3,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.3,0.5,0.7,0.9,1.0 | 0.3          | BEDFAC |
| BEDFAC | bp_1024_e10 | - | 0.15          | ONN_anyV3 |

## ONN_AOM2.fp16.safetensors
![](https://huggingface.co/paimonimpact/ONN/resolve/main/images/oon002.jpg)
▼Use Models
- ONN_anyV3.fp16.safetensors
- Bra6-2(beta).safetensors
- instagram-latest-plus-clip-v6e1_50000.safetensors
- dreamshaper_631BakedVae.safetensors

| Model: A     | Model: B | Weight                                                                | Base alpha | Merge Name          |
| ------------ | -------- | --------------------------------------------------------------------- | ---------- | ------------------- |
| instagram-latest-plus-clip-v6e1_50000 | Bra6-2(beta) | 0,0.0,0.0,0.0,0.0,0.1,0.3,0.5,0.7,0.5,0.3,0.1,0.0,0.0,0.1,0.3,0.5,0.7,0.9,0.7,0.5,0.3,0.1,0.1,0.0,0.0 | 0(cosineB)          | insta_bra |

| Model: A | Model: B | Model: C | Interpolation Method | Merge Name |
| -------- | -------- | -------- | -------------------- |  ---------- |
| insta_bra  | dreamshaper_631BakedVae | v1-5-pruned | Add Difference @ 0.7 |  onn_real     |

| Model: A     | Model: B | Weight                                                                | Base alpha | Merge Name          |
| ------------ | -------- | --------------------------------------------------------------------- | ---------- | ------------------- |
| ONN_anyV3 | onn_real | 0,1.0,0.9,0.7,0.5,0.3,0.1,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.3,0.5,0.7,0.9,1.0 | 0          | ONN_AOM2 |

## ONN_anyV4.fp16.safetensors
![](https://huggingface.co/paimonimpact/ONN/resolve/main/images/oon003.jpg)
▼Use Models
- ACertainty.ckpt
- ONN_AOM2.fp16.safetensors

1. ACertaintyをanyV4.zipのデータセットでFT  = FT_ACertainty
2. ONN_AOM2と単純マージ
| Model: A     | Model: B | Base alpha | Merge Name          |
| ------------ | --------  | ---------- | ------------------- |
| FT_ACertainty | ONN_AOM2 | 0.5          | ONN_anyV4 |

## ONN_pastel.fp16.safetensors
![](https://huggingface.co/paimonimpact/ONN/resolve/main/images/oon004.jpg)
▼Use Models
- bp_1024_e10.ckpt
- ONN_AOM2.fp16.safetensors
- onn_real.fp16.safetensors

1. pastel_A ~ EのデータセットでLoRaを5つ作成 = onnpastelLoRaA~E
2. ONN_AOM2とpastelLoRAをマージ
| Model | Lora | Weight | Merge Name |
| --- | --- | --- | --- |
| ONN_AOM2 | onnpastelLoRaA | 0.8 | onnpastel_baseA |
3. onnpastel_baseAとbp_1024_e10.ckptをマージ
4. | Model: A     | Model: B | Base alpha | Merge Name          |
| ------------ | --------  | ---------- | ------------------- |
| onnpastel_baseA | baseAとbp_1024_e1 | 0.5          | onnpastel_baseB |

5. ACertaintyをany_C.zipのデータセットでFT = onnpastel_baseC
6. onnpastel_baseBとonnpastel_baseCをマージ
| Model: A     | Model: B | Base alpha | Merge Name          |
| ------------ | --------  | ---------- | ------------------- |
| onnpastel_baseB |onnpastel_baseC | 0.5          | onnpastel_baseBC |
7. onnpastel_baseBCとonn_realをマージ
| Model: A     | Model: B | Weight                                                                | Base alpha | Merge Name          |
| ------------ | -------- | --------------------------------------------------------------------- | ---------- | ------------------- |
| onnpastel_baseBC | onn_real | 1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0 | 0          | onnpastel |
8. 各種LoRAをマージ
| Model | Lora | Weight | Merge Name |
| --- | --- | --- | --- |
| onnpastel | onnpastelLoRaB | 0.2 | onnpastel-1 |
| onnpastel-1 | onnpastelLoRaC | 0.3 | onnpastel-2 |
| onnpastel-2 | onnpastelLoRaD | 0.5 | onnpastel-3 |
| onnpastel-3 | onnpastelLoRaE | 0.6 | onnpastel-4 |
| onnpastel-4 | onnpastelLoRaA | 0.2 | onnpastel-5 |

9. 再度マージして微調整
 | Model: A     | Model: B | Weight                                                                | Base alpha | Merge Name          |
| ------------ | -------- | --------------------------------------------------------------------- | ---------- | ------------------- |
| onnpastel-5 | onnpastel_baseC | 0.7,0.5,0.3,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.3,0.5,0.7 | 0          | ONN_pastel |