LTX Video (LTXV) from Lightricks represents a paradigm shift in AI video generation as the first DiT (Diffusion Transformer) based model capable of generating high-quality videos in real-time - faster than it takes to watch them. The model produces stunning 30 FPS videos at 1216×704 resolution.
Breakthrough Multiscale Rendering Technology
In May 2025, Lightricks released LTXV-13B, featuring groundbreaking 'multiscale rendering' technology that delivers both speed and quality through a layered generation process. This innovation makes LTXV approximately 30 times faster than comparable models.
60+ Second Video Generation
The July 2025 update revolutionized the platform by enabling clips longer than 60 seconds - an eightfold leap beyond the industry norm. Live-streamed rendering returns the first second of content almost instantly and continues generating forward in real time.
Ethical Training on Fully-Licensed Data
LTX Video is trained exclusively on fully-licensed data from Getty Images and Shutterstock, ensuring all generated videos are completely free from copyright infringements. This ethical foundation makes it ideal for commercial applications where licensing matters.
Comprehensive Feature Set
- Image-to-video transformation
- Keyframe-based animation control
- Bidirectional video extension (forward and backward)
- Video-to-video transformations
- Open-source with free basic model access
- 30x faster than comparable video models
Implementation Example: Real-Time Text-to-Video Generation
Get started with LTX Video's real-time generation capabilities using the Diffusers library:
Advanced Example: Image-to-Video with Keyframe Control
Transform static images into dynamic videos with precise keyframe-based animation control:
Production Example: 60+ Second Long-Form Content Generation
Leverage LTX Video's 60+ second capability with live streaming rendering for long-form content:
Conclusion
LTX Video combines unprecedented speed (real-time 30 FPS generation), ethical training practices (fully licensed data), and versatile capabilities (image-to-video, keyframe animation, 60+ second clips) to create the most complete open-source video generation solution available in 2025.