All tasks / Models
HunyuanVideo-Avatar
Tencent's open-weights audio-driven model that animates characters from an image into talking or singing videos.
- Tencent
- Open weights
- Released 28 May 2025
Where it ranks
- –Avatars & dubbingModels · Local · no result on this task’s benchmarks yet
Facts
- Vendor
- Tencent
- Family
- HunyuanVideo-Avatar
- Released
- 28 May 2025
- Weights
- Open: download and run it yourself
- GPU memory to run locally
- ~24 GB (README: minimum 24 GB GPU for 704x768, 129 frames (very slow), 96 GB recommended; a 10 GB single-GPU mode is available via Wan2GP)
- 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
- –WanGP (Wan2GP)Apps · Local