HeyGen Launches Avatar 6 with Script-Aware Gestures and 2K Video

Unite.AI
Read full story· 5 min
HeyGen released Avatar 6, an audio-driven avatar model that creates 2K video of a person from a photo and voice recording, featuring script-aware gestures and long-form video generation. It supports detailed motion control, multi-speaker frames, and efficient video editing with minimal frame changes.

What happened

HeyGen launched Avatar 6, a new avatar model, on November 17, 2026. The model creates 2K video of a person from a single photo and an optional voice recording. Avatar 6 is available through HeyGen’s web app, mobile app, and API. It features script-aware gestures, long-form video generation, and supports detailed motion control and multi-speaker frames.

Key details

  • Avatar 6 outputs native 2K video and has no fixed duration limit.
  • Users can specify mood, facial expression, and hand movement for fine control.
  • The model can generate continuous videos up to 28 minutes from a single audio track.
  • HeyGen reports an 84.4% indistinguishability score in external rater tests.
  • API access defaults to 1080p and requires consent verification for footage-based avatars.

Why it matters

Video creators, media companies, and businesses can now generate longer, more customizable avatar videos with higher resolution and more precise motion control. The editing tools allow for targeted changes without redoing entire videos, reducing production time and effort. Consent requirements and API limitations affect how companies can deploy the technology, especially for enterprise users handling realistic video references.


More in Video Generation