Introducing Phoenix
Unparalleled realism, powered by unparalleled research
Tavus’ groundbreaking Phoenix model generates exceptionally realistic talking head videos, complete with natural face movements and expressions accurately synchronized with input.
Developed by our team, this novel approach bypasses traditional methods and constructs dynamic, three-dimensional facial scenes using neural radiance fields (NeRFs).
Using our API, developers can access video generation with unprecedented realism and customization, enabling a wide range of applications.
Access Phoenix using our Replica API.