antgroup/echomimic
Source: https://github.com/antgroup/echomimic
Structured fields
{
"archived": false,
"checked_at": "2026-10-17T00:00:00.000Z",
"usefulness": "Audio-driven (and landmark/pose-editable) portrait animation from Ant Group, AAAI 2025; README names V2 (semi-body) and V3 (unified multi-modal) successor projects.",
"source_type": "repository",
"availability": "available",
"dependencies": [
"CUDA 11.7+",
"Python 3.8/3.10/3.11"
],
"next_check_at": "2026-10-31T00:00:00.000Z",
"hardware_notes": "See evidence.",
"known_limitations": [
"Most recent dated update on the page is 2024-12-10; release-plan items still TBD."
],
"upstream_updated_at": "2026-04-07T04:02:07.000Z"
}Evidence claims
- [upstream_claim] README: 'Lifelike Audio-Driven Portrait Animations through Editable Landmark Conditioning' accepted to AAAI 2025; inputs include audio, facial landmarks, pose or a driver video. source
- [upstream_claim] README names EchoMimicV2 (semi-body human animation) and EchoMimicV3 (1.3B-parameter unified multi-modal, multi-task human animation) as follow-on repositories (their pages were not fetched). source
- [upstream_claim] Tested configurations: A100 80G, RTX 4090D 24G, V100 16G; CUDA 11.7+; accelerated models reportedly about 50 seconds for 240 frames on a V100 versus about 7 minutes before acceleration. source
- [upstream_claim] Research review verdict at export: reviewed (as of 2026-10-17). Recorded by the research pipeline, not a hub review outcome; evidence.review stays unreviewed until an independent reviewer records one.
Rights
| component | license | hosting | redistribution | modification | training | Evidence and notes |
|---|---|---|---|---|---|---|
| code | Apache-2.0 | permitted | permitted | permitted | unknown | Repository page lists Apache-2.0. |
| model_weights | unknown | unknown | unknown | unknown | unknown | No separate weights license on the page; base models (sd-vae-ft-mse, sd-image-variations-diffusers, Whisper tiny) are governed by their own hubs. |