Model · Talking presenters, Video

EchoMimic

A free, downloadable model from Ant Group that makes a photo of a person talk or sing in time with an audio clip. Newer versions animate the upper body and hands too. Runs on a home graphics card with about 12 GB of memory; needs technical setup.

Go to EchoMimic’s website

At a glance

Open or closed: Fully open
Weights and code downloadable under a standard open licence
V3 code and models stated as Apache 2.0. V1 code is Apache 2.0 but its weights licence isn't named.
source · checked 12 Oct 2026 · checked on the maker's own page
Licence: Apache 2.0 (Permissive (e.g. Apache 2.0, MIT))
V3 adds a list of banned uses (illegal, harmful, misinformation, targeting vulnerable groups). V3 is built on Alibaba's Wan2.1-Fun 1.3B, whose own licence also applies.
source · checked 12 Oct 2026 · checked on the maker's own page
Can I sell what I make?: Yes
Apache 2.0 allows commercial use. V1's page says it is meant for academic research, so check V1 weights before commercial use.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial
Cost: Free option: Free to download and run yourself
Plans below.
How you use it: On your own computer, Through another service
Tested on Linux; a Windows one-click package exists for the quantised V3. Weights on Hugging Face and ModelScope; V1 has a Hugging Face demo Space.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial
Skill needed: Technical
Our judgement: Python and CUDA install; V1 has a local web interface.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial
Made by: Ant Group
Built by Alipay's terminal technology team at Ant Group. No company address read, so no country.
source · checked 12 Oct 2026 · checked on the maker's own page
Status: Available
V1 (2024), V2 (half-body) and V3 (Aug 2025) are separate repos; V3-Flash released 22 Jan 2026.
source · checked 12 Oct 2026 · checked on the maker's own page

What you can make with it

A talking presenter, a talking version of a photo (main use)
A video, by bringing a photo to life
A video, lips that match new audio

audio to videoimage to video

Audio-driven portrait and body animation. V1 can also follow facial landmark motion.

source · checked 12 Oct 2026 · checked, but the page was unclear or partial

Plans and prices

No paid plans listed.

source · checked 12 Oct 2026 · checked on the maker's own page

Rights and safety

Real faces and voices: Copies a real face (animate any portrait), no consent step
Self-run software works on any face photo; no consent step. Users are 'fully accountable' for use.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial

Where you can run it

Hugging Face

Other options for “A talking presenter, a talking version of a photo”

Akool, D-ID, DeeVid AI, DomoAI, Dreamina, Dzine, Elai.io, fal, FantasyTalking, Hallo2, Hedra, HeyGen, HunyuanVideo-Avatar, InfiniteTalk, Jogg AI, LivePortrait, Magic Hour, MuseTalk, OmniHuman, SadTalker, VEED, Vidnoz AI, Vozo, WaveSpeedAI

About this entry

Facts last checked 12 Oct 2026 against the maker’s own pages (each fact links to its source). Entry status: draft. Plain-English summary written by Plainly.