Update Notes: Please use both LoRa models together with T2V and I2V. Don't complain to me about it not working or anything like that. If my disciples can use it, it means that if it doesn't work for you, your intelligence might need to be improved by doing more research. Don't blame the model! This is God's choice, I'm not here to complain to you.

Avoid using the ltx-2.3-spatial-upscaler. It completely over-standardizes the video and strips away all its creativity. Just crank up the native resolution directly! The spatial upscaler gives everything this awful plastic look and totally throws off the motion dynamics.

Animation_-v1.0.0:

Description: Training primarily focuses on 3D ultra-, provocative androgynous characters.

Pros: Recognizes a soft lower body, slightly deeper

Cons: Blurry

Improvements: <LORA:0.26~51>

Anime_-1.0.9:

Description: Features exaggerated 2D expressions and ultra-creative, designs!

Pros: High resolution, Creative and with a stronger sense of space

Cons: Distorted and bizarre

Improvement: Use with two LORA units