All tasks / Models
MuseTalk 1.5
Open-weights real-time lip-sync model that re-renders the mouth region of a face video to match input audio.
- Tencent Music (Lyra Lab)
- Open weights
- Released 28 Mar 2025
Where it ranks
- –Avatars & dubbingModels · Local · no result on this task’s benchmarks yet
Facts
- Vendor
- Tencent Music (Lyra Lab)
- Family
- MuseTalk
- Released
- 28 Mar 2025
- Weights
- Open: download and run it yourself
- License
- MIT (code); model weights usable for any purpose incl. commercial
- GPU memory to run locally
- ~4 GB (README: tested on an RTX 3050 Ti laptop GPU with 4 GB VRAM in fp16 (8 s video in about 5 min); 30 fps+ real-time on a Tesla V100)
- Access
- Free (open weights)
- Last verified
- 18 Oct 2026
Price from github.com, 18 Oct 2026.
Benchmark results
No public benchmark results recorded yet.
Where you can use it
No tracked tool lists this model yet; as open weights it runs in local runtimes.
Earlier versions
- MuseTalk 1.0 · 2 Apr 2024