#7070th of 75 in Image and videoFor image, video and design, video and animation
echomimic_v2
Audio-driven half-body human animation from a reference image and pose sequence
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 4.7k
- License
- Apache-2.0
- Last commit
- Feb 2026
- Language
- Python
4,489 on 22 Apr 20264,678 on 11 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
EchoMimicV2 generates talking half-body human videos from a reference image, an audio clip and a pose sequence. It is the CVPR 2025 research code from Ant Group, with pretrained English and Mandarin checkpoints on Hugging Face and ModelScope. It ships a Gradio app, inference scripts, an accelerated variant and the EMTD dataset processing scripts.
Who it is for: Researchers and engineers testing audio-driven avatar video on NVIDIA GPUs
Strengths
- Accelerated inference script cuts 120 frames from ~7 min to ~50 s on an A100
- Gradio UI (app.py, app_acc.py) plus a ComfyUI node contributed by the community
- Pretrained checkpoints cover English and Mandarin audio, hosted on Hugging Face and ModelScope
- Apache-2.0 code with a published paper and dataset processing scripts
Weaknesses
- Requires an NVIDIA GPU with CUDA >= 11.7; tested only on 16 GB to 80 GB cards
- Setup is manual: conda, pinned torch 2.5.1, static ffmpeg and separate weight downloads
- Stated as academic research only; no hosted demo, the online demos are still TBD
- No Docker image and no tagged releases
What it needs
- GPU required
- Needs CUDA >= 11.7, Python 3.10, PyTorch 2.5.1, xformers, ffmpeg-static 4.4, git-lfs, Whisper tiny audio processor, sd-vae-ft-mse
- Models: EchoMimicV2 checkpoints (denoising_unet, reference_unet, motion_module, pose_encoder), sd-vae-ft-mse, Whisper tiny
Also in Image and video
See all 75| Rank | Project | Score |
|---|---|---|
| 1 | frigateLocal NVR with realtime object detection for IP cameras | 93 out of 100 |
| 2 | MoneyPrinterTurboGenerates short videos from a topic, with script, footage, voiceover and subtitles | 88 out of 100 |
| 3 | toonflow-appInfinite-canvas tool for AI short drama and comic video production | 87 out of 100 |
| #4 | InvokeAIWeb UI for diffusion image generation with inpainting canvas and node workflows | 86 out of 100 |
| #5 | ComfyUINode-graph editor for running diffusion image, video, audio and 3D workflows | 85 out of 100 |