About this model
LTX Video 2.0 19B is the text-to-video configuration of Lightricks' LTX-2 generation, an open-source, DiT-based audio-video model that produces synchronized video and audio in a single unified pass. Lightricks released the model weights publicly through its Hugging Face repository. The model is paired with a faster LTX Video 2.0 19B Distilled checkpoint and an image-to-video variant.
Relative to the earlier LTX Video 2.0 Full Quality family, this release adds native, single-model audio generation alongside video. It keeps the iterative, lower-latency design that characterizes the LTX line.
Lightricks has continued iterating: the later LTX Video 2.3 Full Quality arrived in March 2026, following this January 2026 release. The model supports both API access and local ComfyUI deployment.
This About section is AI-generated from public sources (Claude Opus 4.8), with no human editing. It may contain inaccuracies — verify critical details against the sources listed above.
Data sources: Venice API · HuggingFace · Wikipedia — enrichment updated 1d ago