Atmee v2: an invite-only real-time avatar video model

Chen-Ping Yu (Atmanity) launched Atmee v2, a real-time video foundation model for face-to-face conversational avatars, plus Causeur as the context 'brain'. Avatars from one image and about 10 seconds of audio; Web, iOS, Android and kiosk; SDK setup by pasting a markdown prompt into a coding agent. Speed and cost figures (for example a 10-second clip in 3.5s for under $0.01) are the company's own. The demos are staged motion-graphic mock-ups, access is invite-only at atmee.ai, and there's nothing readers can test yet. The founder on camera looks genuine.

Why it mattersA serious founder pitch for face-to-face AI interfaces, but treat the speed/cost race and product footage as marketing until invitees can reproduce them. We haven't tested it hands-on.

Curated by Sara Simeone as part of our human editorial coverage. The original sources are linked below; curation is distinct from a Hype or Fact? assessment.

SHARE THISPost on X ↗Share on LinkedIn ↗Copy link

How we knowCompany figures are labelled as their own claims; anything we couldn't verify is labelled as a report or a claim.