Tavus Griffin: the first Human Interaction Model
Tavus launched Griffin, a full-duplex video-to-video model that listens, watches, and generates expressive speech and video in real time, claiming 48% of blind-study participants thought they were talking to a human.
Why it matters
First claim of a real-time video Turing-test pass (48% vs 2.4% for its previous stack) and 3.83 on NVIDIA's VideoFDB against a 3.92 human reference — defines a new 'Human Interaction Model' category beyond turn-based avatars.
What you could build with it
A telehealth startup builds an always-on video check-in companion that notices when a patient looks confused or pauses mid-sentence and adapts its tone and pacing in response, resulting in noticeably higher completion rates for at-home rehab exercises.
Does it hold up?
Too early to judge: Griffin-Lite is a research preview for select testers only, and the 48% video-Turing-test and VideoFDB 3.83 numbers come from Tavus's own study — independent verification of latency and quality has not been published.
Built with Tavus Griffin
- THE DECODER: Nearly half of test subjects mistook Tavus' AI video avatar for a real personarticle · Independent coverage detailing Griffin's architecture (replaces chained Phoenix/Sparrow/Raven pipeline) and noting the 48% figure comes from Tavus's own 54-person study.
Learn more
First spotted on article: source.
More AI models
ElevenLabs Eleven v4 + v4 Turbo: new emotive TTS architecture with a 100ms real-time variantElevenLabs released Eleven v4 and Eleven v4 Turbo on September 28: a new text-to-speech architecture with inline…models · JEV 0.72VoiceStudio: open-source, fully-local ElevenLabs alternative with voice cloning, dubbing and dictation in 646 languagesVoiceStudio is an open-source, fully-local voice AI studio: voice cloning from a 3-second sample, voice design, video…models · JEV 0.71DeepSeek V4-Flash official API: public beta with upgraded agent capabilities, Responses API and Codex supportDeepSeek's official V4-Flash is now a public beta API with massively upgraded agent capabilities - benchmark scores…models · JEV 0.69Mistral Forge: enterprise platform for training and continuously improving proprietary modelsMistral AI launched Forge (Sep 28) — a full-lifecycle model training platform (pre-training, SFT, DPO/ODPO, RL…models · JEV 0.69kev: open, trainable Jev-like family of small decision models on Qwen3.5/3.8A family of small decision models (0.8B to 27B) built on Qwen3.5/3.8 that reproduces Jev's prefill-only typed-decision…models · JEV 0.68GPT-Synopsys: OpenAI and Synopsys multi-year deal to build an AI model that operates EDA toolsOpenAI and Synopsys signed a multi-year agreement on Sep 30 to jointly develop GPT-Synopsys, a specialized model that…models · JEV 0.66
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.