The package covers text-to-video, text-to-audio, image-to-video, video extension, audio extension, video inpainting, audio inpainting, video outpainting, IC-LoRA references, audio-to-video, and video-to-audio tasks. Its documentation includes quick start, dataset preparation, training modes, configuration, training, inference, utilities, and troubleshooting guides.
LTX-2 Trainer is useful for teams that want to adapt an audio-video foundation model to custom datasets, styles, or references. It expects local LTX-2 checkpoints, a Gemma text encoder, Linux with CUDA, and substantial GPU memory for standard configurations.
