LTX-2: Real-Time-Class Video Generation
LTX-2 is Lightricks' DiT-based video model; the 13B distilled variant is optimized for speed, producing 5–10 second clips in well under a minute, sometimes in seconds. It supports image-to-video with synchronized audio in some configurations and outputs up to 1080p. Quality is solid for its speed and price, which is why youcut.ai positions it as the fast/draft tier at 6 credits.
Strengths
- • Fastest generation in the lineup
- • Cheapest: 6 credits per clip
- • Up to 1080p
- • Open weights available
Limitations
- • Less complex motion than Kling or Seedance
- • Occasional temporal flicker on fine detail
- • Prompt adherence weaker for multi-subject scenes
Best for
- • Rapid iteration on prompts
- • Simple camera moves and ambient motion
- • Budget social content
| Specs | |
|---|---|
| Parameters | 13B (distilled) |
| Resolution | Up to 1080p |
| Duration | 5–10 s |
| Latency | ~10–60 s |
| Provider key | fal-ai/ltx-video-13b-distilled/image-to-video |
Try LTX-2 (13B distilled) in
Compare
Frequently asked questions
Why is LTX-2 so fast?+
The distilled model needs far fewer denoising steps and uses a highly compressed video latent space.
How much does it cost?+
6 credits per clip, about $0.18 on the Pro plan, the cheapest video option on youcut.ai.
Is quality good enough to publish?+
For simple scenes and social posts, yes. For complex action, draft with LTX-2 and render the final with Kling or Seedance.
Does it generate audio?+
Some LTX-2 configurations do; the distilled image-to-video endpoint on youcut.ai currently outputs silent video.
Is it open source?+
Lightricks publishes open weights for LTX models with a permissive license for most uses.