Atmee v2 turns one photo and 10s of voice into a live video-callable AI avatar
rohanpaul_ai · x · 2026-11-05
Atmanity launched Atmee v2, a real-time avatar model that turns 1 photo + 10 seconds of someone's voice into a talking character you can video-call live. A dual-model setup — Atmee v2 renders the face and body while a second model, Causeur, drives contextual expressions and gestures. Founder Yuchen Ping (ex-AR startup acquired by Google) frames live video as the next interface for AI agents after text and voice. Available via API with an invite code from atmee.ai.
More from coding & agent
- Context sovereignty: treat every agent input as a stranger — gerardsans · 2026-11-05
- Working with agents: people own decisions, rules belong in code not prompts — gerardsans · 2026-11-05
- Pacing agent pipelines to human review: evidence first, fail soft — gerardsans · 2026-11-05
- Beyond harness engineering: from 2027 factory engineering to 2028 slow engineering — gerardsans · 2026-11-05
- Bridge launches on Product Hunt: video calls rethought for AI agent teammates — thisiskp_ · 2026-11-05
- Render launches Sandboxes: isolated Firecracker microVMs with persistent state for every AI agent — adityaag · 2026-11-05