videoimage
LTX 2.3
LTX-2.3 video across audio-sync, extend, retake, reframe, and fast image and text endpoints to 20 seconds.
Studio is in early access; availability resolves per project after sign-in.
- Endpoints
- 8
- Eligible
- 7
- Input schemas imported
- 8 / 8
- Delivery
- fal.ai
- Weights
- Not stated
- Developer
- Not stated in source
Overview
LTX-2.3 generates up to 20 seconds of video from audio, images, or text. Audio inputs get timing-accurate synced visuals; extend continues clips seamlessly and retake regenerates scene variations. Fast image and text tiers optimize for speed, with 9:16 portrait alongside landscape. A reframe endpoint is also listed.
Capabilities
- Audio-synced generation
- Seamless clip extension
- Scene retake variations
- 20-second i2v and t2v plus fast tiers
Best for
- Synced music visuals
- Extended sequences
- Fast-turnaround drafts
Use cases
- Portrait social video
- Sharper VAE detail passes
- Prompt-tuned compositions
Endpoints
7 eligible · 1 under review. Cataloged is not the same as executable: Studio resolves which endpoints a project can run.
| Endpoint | Task | Catalog status | Inputs |
|---|---|---|---|
fal-ai/ltx-2.3/audio-to-video | video editing | Eligible | 5 fields · requires audio_url |
fal-ai/ltx-2.3/extend-video | video editing | Eligible | 5 fields · requires video_url |
fal-ai/ltx-2.3/image-to-video | image to video | Eligible | 8 fields · requires prompt, image_url |
fal-ai/ltx-2.3/image-to-video/fast | image to video | Eligible | 8 fields · requires prompt, image_url |
fal-ai/ltx-2.3/reframe | unknown | Under review | 3 fields · requires video_url |
fal-ai/ltx-2.3/retake-video | video editing | Eligible | 5 fields · requires video_url, prompt |
fal-ai/ltx-2.3/text-to-video | text to video | Eligible | 6 fields · requires prompt |
fal-ai/ltx-2.3/text-to-video/fast | text to video | Eligible | 6 fields · requires prompt |
Questions
Max length?
Up to 20 seconds from image or text.
Extend versus retake?
Extend continues clips; retake regenerates variations.