Lora hashes: "shimabara_yuuhi_neo_ranga_anima-preview3_lora_20260414_ep-000020: ae322e87aa22"

This is a multi-character LoRA model featuring Yuuhi from Neo Ranga as the main character, designed for anima preview3.

Note: Due to the training data, no mouth outputs appear frequently. This will need to be addressed in future tag updates. For now, mitigate it by adding "no mouth" to the negative prompt.

A LoRA weight around 0.8 is recommended. At 1.0, poses tend to break, while at 0.6, fidelity drops. Adjust up or down depending on your use case.

What "_no_te" ver means: this is a version with the TE trained component removed afterward. Output quality remains the same, and the LoRA file size is smaller, so this version may be sufficient.

You should use shimabara yuuhi to trigger the image generation.

You should use joel \(neo ranga\) to trigger the image generation.

You should use shimabara ushio to trigger the image generation.

You should use shimabara minami to trigger the image generation.

Download them in the Files & versions tab.