v3/v2-real

Recommend Prompt

score_9, score_8_up, score_7_up, source_real, real_skin,
...what you want here...

Recommend Negative Prompt

score_6, score_5, score_4, source_cartoon, source_anime, source_pony, source_furry,

worst quality:1.2, low quality:1.2, normal quality:1.2, disabled body, sketches, (manicure:1.2), lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, extra fingers, fewer digits, cropped, jpeg artifacts, signature, watermark, username, blurry, (ugly)

這個版本融合了自行訓練的 Pony LoRA/LoCon ，會更偏向真實表現。

This version incorporates self-trained Pony LoRA/LoCon, which will be more realistic.

このバージョンには、自己トレーニングされた Pony LoRA/LoCon が組み込まれており、よりリアルになります。

이 버전에는 자가 훈련된 포니 LoRA/LoCon이 포함되어 있어 더욱 현실감이 커집니다.

v1-rev4

這是一個融合模型，使用了自行訓練的 Pony LoRA/LoCon 與以下模型進行融合：

1. tPonynai3

2. DucHaiten-Pony-XL (no-score)

所使用的 Lora/LoCon 為自行訓練的模型：

1. Pony Realistic Tweaker

2. Maybe Better Pose Lora XL

3. MaybeBetterFaceInFullBodyLora_XL_rev3

4. hinapony_realism

5. hinapony_more_realistic_tweaker

模型為了保留 Pony 的特性，並沒有過渡強化真實效果，使用上也不一定需要 Pony 的評分標籤 score_* ，但有使用的話也確實能增加畫面的穩定性。

This is a model that uses a self-trained Pony LoRA/LoCon to merge with the following models:

1. tPonynai3

2. DucHaiten-Pony-XL (no-score)

The Lora/LoCon model used is a self-trained model:

1. Pony Realistic Tweaker

2. Maybe Better Pose Lora XL

3. MaybeBetterFaceInFullBodyLora_XL_rev3

4. hinapony_realism

5. hinapony_more_realistic_tweaker

In order to preserve the characteristics of Pony, the model does not over-enhance the real effect, and the use of Pony's rating labels score_* is not necessarily required, but it does increase the stability of the screen if used.

これは、自分で訓練したポニーLoRA/LoConを使って、以下のモデルと融合させるモデルである：

1. tPonynai3

2. DucHaiten-Pony-XL (no-score)

使用した Lora/LoCon モデルは、自己学習モデルである：

1. Pony Realistic Tweaker

2. Maybe Better Pose Lora XL

3. MaybeBetterFaceInFullBodyLora_XL_rev3

4. hinapony_realism

5. hinapony_more_realistic_tweaker

ポニーの特性を生かすため、実効果を過剰に強調しないモデルとなっており、ポニーのレーティングラベル score_* の使用は必ずしも必要ではないが、使用すれば画面の安定感が増す。

자체 훈련된 포니 LoRA/LoCon 을 사용하여 다음 모델과 병합하는 모델입니다:

1. tPonynai3

2. DucHaiten-Pony-XL (no-score)

사용된 Lora/LoCon 모델은 자체 학습된 모델입니다:

1. Pony Realistic Tweaker

2. Maybe Better Pose Lora XLMaybeBetterFaceInFullBodyLora_XL_rev3

3. hinapony_realism

4. hinapony_more_realistic_tweaker

포니의 특성을 유지하기 위해 모델은 실제 효과를 과도하게 향상시키지 않으며, 포니의 등급 라벨 score_* 을 반드시 사용해야 하는 것은 아니지만 사용하면 화면의 안정성을 높일 수 있습니다.