HeyGen Launches Avatar 6 AI Avatar Model
The audio-driven model generates gestures from speech context and delivery, addressing repetitive motion in AI avatars.
Updated on Nov. 18, 2026 in Artificial Intelligence

Live Poll
Would you trust a video of someone talking to you if they were AI-generated?
HeyGen has launched Avatar 6, an audio-driven AI avatar model that generates gestures based on the context and delivery of speech. Users can create an avatar from as little as 30 seconds of video, and third-party platforms can now generate photo-based and virtual avatars with the model.
Why it matters
The model targets repetitive motion, a persistent limitation of AI avatar technology, and HeyGen says realistic videos of real people remain within its platform, where explicit consent is required. The launch follows HeyGen's expansion to more than 40 million users across 196 countries.
Avatar 6 needs as little as 30 seconds of source video per user, and 84% of participants in Turing-style testing rated it as equally realistic as the person it represents. HeyGen has not published the test sample size or the competing models in the Avatar Arena ranking.
The players
HeyGen
HeyGen is an AI video company whose platform lets users create avatars and videos, with reported reach across 196 countries.
The details
The model interprets a user's script and generates corresponding gestures, using footage of each individual to model how they naturally look, move and express themselves. Third-party platforms can now generate photo-based and virtual avatars with Avatar 6 through the HeyGen API.
Timeline
June: HeyGen surpassed $200 million in annual recurring revenue.
November 17, 2026: HeyGen launched Avatar 6 through its platform and API.
The Tech Race
Avatar 6 extends HeyGen's earlier HeyGen Video and HeyGen Voice releases into gesture-driven avatars, and the company's Voice product debuted at #1 on Artificial Analysis' leaderboard. The launch places HeyGen among competing AI video and avatar providers that are racing to make synthetic presenters less repetitive and more realistic.
Creators and small business owners can make a usable avatar from about 30 seconds of video, and the survey found 71.6% of small business owners had recorded a business video without posting it. The tool may lower the effort needed to produce regular video content.
The takeaway
Consider whether short audio-driven avatar videos could replace recording tasks you have been putting off, but review platform consent rules before using a real person's likeness.
Further reading
For broader context on how AI avatar and video tools are developing, see Artificial Intelligence coverage.
Source note: This article includes information reported by Longview Daily News.
Live Poll
Would you trust a video of someone talking to you if they were AI-generated?










