LTX-2.3
Video modelPrevious generationOpen weightsCard summarizedA DiT-based audio-video foundation model designed to generate synchronized video and audio within a single model.
Source: LTX-2.3 Model Card
Where to call it
- LTX API
ltx-2-3-pro - Self-host (weights)
Lightricks/LTX-2.3
Good for
- Synchronized video and audio in one pass
- Local execution and fine-tuning (LoRA, IC-LoRA, full)
- Image-to-video with more natural motion
Watch out for
Overview
Lightricks · 01LTX, front
LTX-2.3
A DiT-based audio-video foundation model designed to generate synchronized video and audio within a single model.
Typical price
$0.08 / sec
Pro, 1080p, text- or image-to-video
- Max resolution
- 4K (3840×2160 / 2160×3840)
- Max length
- 20 s (Fast at 720p/1080p, 24/25 fps); 10 s otherwise
- Modes
- text-to-video, image-to-video
- Released
- 5 Mar 2026
LTX-2.3 is a significant update to LTX-2, a DiT-based audio-video foundation model that generates synchronized video and audio in one model, with open weights and a focus on local execution. Lightricks cites better audio and visual quality and stronger prompt adherence; the API launch added a new VAE latent space, cleaner audio, last-frame interpolation, portrait video and 24/48 fps. Model card intro; API changelog 5 Mar 2026
ltx-2-3-pro- Typical price
- $0.08 / sec
- Max resolution
- 4K (3840×2160 / 2160×3840)
- Max length
- 20 s (Fast at 720p/1080p, 24/25 fps); 10 s otherwise
- Aspect ratios
- 16:9, 9:16
- Frame rate
- 24, 25, 48, 50
- Released
- 5 Mar 2026
- text-to-video
- image-to-video
- first and last frame
- audio-to-video
- retake (video editing)
- extend
- HDR upscale