Lightricks LTXV
Open, production-grade multimodal video-generation model (LTX-2.3) with synced audio, deployable via API, on-prem or open weights.
What it does
LTX (by Lightricks) is a multimodal video-generation foundation model. Its latest LTX-2.3 is a 22B-parameter dual-stream diffusion transformer that jointly generates video and audio, with native portrait output, HDR and up to 20-second clips. It ships open source (free for companies under $10M revenue) and is deployable via API, on-prem or open weights, with tools like LTX Playground, Desktop and Trainer.
Core features
LTX-2.3 joint audio-video generation
Text-to-video and image-to-video
Native portrait output up to 1080x1920
HDR output and up to 20-second clips
Open weights, API and on-prem deployment
LTX Trainer for custom characters and styles
Best for
→Generate production video from text or images
→Build AI video features into products via API
→Fine-tune the model on custom IP and styles
→Run local video generation with LTX Desktop
Toolspool rankingby global site rank