Had a chance to test the new Tavus model that set the Internet aflame. I did not expect the avatar ...
TL;DR · AI 摘要
Justine Moore on X: "Had a chance to test the new Tavus model that set the Internet aflame. I did not expect the avatar...
核心要点
- 主题聚焦：Had a chance to test the new Tavus model that se
- 来源：Justine Moore(@venturetwins)，建议结合原文判断细节。
- AI 分析暂不可用，本条为保底评分与摘要。
Justine Moore on X: "Had a chance to test the new Tavus model that set the Internet aflame. I did not expect the avatar to so effectively roast my job ☠️ Impressive progress in real-time interactive video - this is without any cuts or speeding up clips! Some of the mannerisms were shockingly real… / X
Justine Moore
@venturetwins
Had a chance to test the new Tavus model that set the Internet aflame. I did not expect the avatar to so effectively roast my job ☠️ Impressive progress in real-time interactive video - this is without any cuts or speeding up clips! Some of the mannerisms were shockingly real.
$
00:00
/$
Tavus
@tavus
Oct 1
Introducing Griffin, the first model to pass the video Turing test. 48% of people who talked to it live thought it was a real human. Previous systems have had a pass rate <3%. It is #1 on NVIDIA's benchmark for full-duplex AI video. It’s the first Human Interaction Model (HIM).
Readers added context
The 48% figure and "video Turing test" claim are from Tavus's own study of 54 one-minute calls, not independently verified or using a standard protocol. Griffin-Lite leads NVIDIA's VideoFDB benchmark on their public leaderboard.
cellcog.ai/blog/tavus-gri…
research.nvidia.com/labs/amri/proj…
tech-ish.com/2026/10/02/tav…
7:01 PM · Oct 6, 2026
·
9,832
Views
21
6
82
30