techfreakworm/ltx2.3-aio-generator

LTX 2.3 Studio is a web-based application that lets you generate videos from text prompts, images, or audio using an open-source 22-billion-parameter AI video generation model called LTX-2.3, made by Lightricks. Think of it as a single interface that unlocks six different ways to create or transform video with AI. The six modes cover: generating a video purely from a text description, animating a still image with a prompt, generating video that moves in sync with an audio track, lipsync, making a person in an image mouth words to match an audio clip, keyframe interpolation, where you supply a first and last frame and the model fills in the motion between them, and style transfer, where the motion from a source video is preserved but the visual appearance is restyled to match a reference image. Each mode offers Fast, Balanced, and Quality presets that control how many generation steps are run, trading speed against output quality. The underlying AI processing is handled by ComfyUI, a node-based image and video generation engine, which LTX 2.3 Studio wraps with a simplified Gradio web interface so you never have to interact with ComfyUI's technical node graph directly. You can run it locally if you have a machine with a large amount of GPU memory (the model weights alone are roughly 70-80 GB), on Apple Silicon hardware, or deploy it to Hugging Face Spaces for cloud-hosted GPU access. The backend is Python 3.11. The full README is longer than what was provided.