All tasks / Models
EchoMimicV3
Open-weights 1.3B audio-driven human animation model for talking portraits and half-body avatars.
- Ant Group
- Open weights
- Released Aug 2025
Where it ranks
- –Avatars & dubbingModels · Local · no result on this task’s benchmarks yet
Facts
- Vendor
- Ant Group
- Family
- EchoMimic
- Released
- Aug 2025
- Weights
- Open: download and run it yourself
- License
- Apache-2.0
- Parameters
- 1.3B
- GPU memory to run locally
- ~12 GB (README: 12 GB VRAM with the provided Gradio UI; 16 GB via ComfyUI nodes)
- Access
- Free (open weights)
- Last verified
- 19 Oct 2026
Price from github.com, 19 Oct 2026.
Benchmark results
No public benchmark results recorded yet.
Where you can use it
- –ComfyUIWorkflow platforms · Local
Earlier versions
- EchoMimicV2 · Nov 2024
- EchoMimic · Jul 2024