Event ·Multi-source

Griffin模型通过视频图灵测试

Tavus团队开发的Griffin模型让48%的实时对话者误以为是真人,远超同类系统不足3%的通过率。该模型在NVIDIA全双工AI视频基准(VideoFDB)的感知与生成两项均排名第一。Griffin能实现全双工对话,在说话的同时持续倾听和观看,并生成整个画面而非仅换脸。

Current read

Tavus的Griffin让近半人误认为真人对话,全双工技术实现实时互动,远超同类系统表现。

9 sources83° AI heatLast updated 2026/10/2 00:32:50

Evidence

9 sources
Related source 1rohanpaul_ai
View source
Related source 4TestingCatalog
View source
Related source 5@koltregaskes
View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.