宫本丽（みやもと れい，Miyamoto Rei）

毒岛冴子（ぶすじま さえこ，Busujima Saeko）

高城沙耶（たかぎ さや，Takagi Saya）

鞠川静香（まりかわ しずか，Marikawa Shizuka）

使用罗马音作为触发词。必须加上触发画得才比较像。

在Animagine3.1模型上训练，并不合适pony模型。

你可以加入 anime coloring tag，但是有时候会比较模糊。需要高清修复。

如果有时候衣服画得不精确，那是你的tag写得不够详细。

关于色彩问题，因为数据集大多来自动画，人物经常处于黄昏和夜晚的场景，因此训练出来后色彩有时候会过渡发黄。这时候需要降低strength或者切换采样器。

Use Romanized words as trigger phrases. You must add the trigger for it to look more accurate.

Trained on the Animagine3.1 model, it is not suitable for the pony model.

You can add the anime coloring tag, but it can sometimes appear blurry. High-resolution fix is needed.

If the clothes are sometimes not drawn accurately, it means your tags are not detailed enough.

Regarding color issues, the dataset comes from anime screenshots, many of which are dusk and nighttime scenes. As a result, the colors can sometimes be unstable after training into the LoRA. In this case, you need to reduce the strength or switch the sampler.