Side by side
See the difference.
Scores, capabilities and costs for the models you’re considering.
Scroll to compare models →
| 2 models | fal-aiLatentSyncRemove × | alibabaAlibaba: Wan 3.0Remove × |
|---|---|---|
| What’s different | LatentSync is a video-to-video model that generates lip sync animations from audio using advanced algorithms for high-quality synchronization. | Can choose clip duration from the prompt; toggle generated audio, prompt expansion and extra thinking before rendering. |
| Date | Mar 25, 2025Listed | Aug 19, 2026Released |
API prices USD · provider configuration shown below | ||
| Listed rate | See provider quotes | $0.05/ secondDeepInfra · all routes ↗480P · $ cost per second · output second |
| Provider endpoints | 0 endpoints → | 2 endpoints → |
Benchmarks Each result retains its evaluation configuration | ||
| AA Video + Audio ↗2.1 | Not reported | 1,174 ↗Text-to-video · audio evaluated · all categoriesIndependent result95% CI 1,159–1,189 · n=5,861 |
| AA Video Silent ↗2.1 | Not reported | 1,135.19 ↗Text-to-video · silent evaluation · all categoriesIndependent result95% CI 1,116.19–1,154.19 · n=3,890 |
Specifications | ||
| Context window | — | — |
| Max output | — | — |
| Parameters | — | — |
| Structured output | Not reported | Supported |
| Image input | Not reported | Supported |
| Audio input | Not reported | Supported |
| Video input | Not reported | Supported |