Griffin: The First Human Interaction Model | Tavus

Griffin is the first Human Interaction Model (HIM). On live video calls, 48% of people who talked to it thought it was a real person.

Introducing Griffin

The First Human Interaction Model

Griffin is the world’s first Human Interaction Model (HIM), a new class of model designed to understand and generate face-to-face real-time human interaction. It listens while they talk, and they pay attention to expressions and pauses, not just words. Griffin combines our earlier research into real-time perception, conversation and expressive human-like video into one unified video to video system.

48%

of people believed Griffin was a real person after a one-minute video call.

#1

on NVIDIA’s independent test of face-to-face AI

37%

ahead of the next best AI at reacting in the moment

By: Hassaan Raza, Co-founder & CEO

Ioannis Patras, Head of Research

Tavus Research Team

Today we're introducing Griffin, our first Human Interaction Model (HIM). HIMs are a new class of models designed to understand and generate face-to-face real-time human interaction. They listen while they talk, and they pay attention to expressions and pauses, not just words.

Griffin builds on our earlier research into real-time perception, conversation and expressive human-like video. It combines those model capabilities into one, unified video-to-video system.

Griffin is the first model to pass the real-time, video Turing test. In a live study, 48% of participants who talked with Griffin thought they had talked with a real human. Previous systems, including Tavus’s own industry leading conversational video interface (powered by Phoenix 4.5, Sparrow-2, Raven-1), had a max 2% pass rate. This represents a huge breakthrough in human-machine understanding, and is only possible due to Griffin’s video-to-video duplex approach with audiovisual generation, movement, conversational modeling – all in one system, operating in real time.

Griffin-Lite, a research preview, is available today to a select group of early testers, with a wider release of a more powerful model to follow. This post covers what Griffin can do, how we built it, how we measured it, our approach to safety and what comes next.

A message from Hassaan:

Humans are evolutionarily designed to communicate face to face. We speak as much through our words as we do our expressions, tone, gestures and timing. So much of human communication is non-verbal. It’s an art, a dance.

Machines don’t understand this art. They require us to meet them where they are, to learn their language- the right command, the right button click, the right prompt.

At Tavus, we believe in a future where machines meet us where we are. Where machines learn to communicate in the way that we most effortlessly do. They learn to see, hear, respond and even look like we do. In effect, we want computing to become invisible. We want it to feel second nature.

In a great conversation, you don’t spend your time evaluating the mechanics. You think about what you want to say, what you're hearing and seeing, and ultimately the conversation just flows. In great conversations, you aren't thinking about the conversation at all.

With machines, those mechanics are distracting and demand your attention.

Did it hear me? Did it understand me? You share something personal or difficult, and the face continues to smile, or the faceless, cold machine sits there, then responds with a monotonous 'I'm sorry to hear that'. You pause to collect your thoughts, and it jumps on and responds to incomplete context. It needs time to answer, but nothing in its expression or voice or gesture tells you it's working on a response.

Each of these moments is you managing the machine. You adjust your timing, simplify your language, make sure to think complete thoughts before saying anything. You remind yourself: don't leave dead air! Remember to be very clear about what you want! You lower your expectations. The system, in all its intelligence, might be capable of infinite incredible work. It can discover new mathematics dammit! Yet, getting the help for the simple stuff takes more effort than you have to give.

Human communication is an art, a dance. The right move, at the right time. A deep understanding of what you mean, when the words themselves could mean many things. A nod of acknowledgment to signal understanding. An expression that tells you everything without saying anything.

All of these are things we do as humans without thinking about it. These signals let your attention stay on the conversation, not how to converse. As a machine becomes more coherent at expressing and understanding these signals, communicating with it, too, becomes something you don't have to think about.

01

Griffin-Lite is a preview of our first Human Interaction Model: a full-duplex video-to-video model that responds to human behavior in real time. Perception, deciding when and how to respond, and expressive speech and video generation all happen at the same time.

To see what it looks like in practice, we brought people who had never used Griffin together with Tavus researchers for open-ended conversations and short demonstrations.

Behavior and emotion modelling

Griffin is able to express behaviors, emotions, and gestures according to conversational context. It laughs, changes its tone, and shifts in response to what the other person says and does.

Rudy gets a promotion.

Brittney and the Buckeyes.

Clip: Rudy gets a promotion.

Full-scene generation

Griffin generates every pixel in every frame in real time from one reference image. It controls the whole scene, not just the face, arms, and fingers, but also things like the movement of the chair the person is sitting in, the shadows they cast, and the background behind them.

Mars plays Simon Says.

Brittney plays Simon Says.

Clip: Mars plays Simon Says.

Full-duplex conversation

Griffin continuously evaluates the state of the conversation and independently of conversational turns. It can interrupt, adjust, back-channel, or be interrupted without losing its place.

Ari tells a story.

Mars talks about a date.

Clip: Ari tells a story.

Perception

Griffin elevates dialogue beyond just verbal understanding, comprehending and reacting to visual context when appropriate.

Ari solves a Rubik's cube.

Mars shows a peacock.

Brittney looks cozy.

Clip: Ari solves a Rubik's cube.

Temporal understanding

Griffin understands time as part of the conversation: how long a silence has lasted, what a sustained silence means, and when to speak again on its own.

Sagar solders a board.

Clip: Sagar solders a board.

02

Technical approach: Building a full-duplex, video-to-video interaction model

Perception, conversational decision-making, and generation are in continuous interplay throughout human-to-human interaction, and we designed Griffin around the same principle. Central to our approach is a conversational model that controls speech and nonverbal behavior directly, with streaming speech and video generators that follow its decisions as they change. Griffin, is a two part system:

1. A Continuous Conversational Modeling engine that perceives the incoming audio and video, decides when and how to respond, and produces signals that determine what should be said and how.
2. An Audio Visual Generation engine comprising a Streaming Speech Generation and a Streaming Video Generation architecture that convert those signals into speech and video, respectively.

Perception, decision-making, and generation run concurrently throughout the conversation, so the model can respond to changes in the conversation while it’s speaking as well as while it’s listening.

Griffin: A Human Interaction Model

Text tokenControl signalAudio tokenVisual tokenShared processing

PAL system prompt

TTTTTT

User video input

VVVVV

User audio input

AAAAA

Full-Duplex Continuous Conversational Modelling

Listen continuouslyProcess audio and video in real time, without interruption.

Turn-taking: speak / hold / yieldDecide when to speak, hold, or yield the floor naturally.

Barge-in: continue / pause / stopRespond to user interruptions or changes.

Backchannels & non-verbal cuesGenerate natural acknowl­edge­ments, expressions and gestures.

Active listening during speechKeep listening and understanding while speaking.

Low-latency streaming coordinationAlign audio, video and conversation flow in real time.

Verbal/non-verbal tokens and controls

Reference audio

Reference image

Streaming Audio-Visual Generation

Audio latents

Video latents

A1A2A3A4A5A6A7A8

V1V2V3V4V5V6V7V8

Rendering

Streaming Audio-Visual Decoding

Output audio & video

Continuous Conversational Modeling: (Conversational) intelligence, Perception and Control

Griffin makes conversational decisions at regular sub-second intervals rather than once per turn. At each interval it assesses the state of the conversation, including what has been said and the user's verbal and nonverbal behavior, and decides what to say next and how to say it. This differs from cascade systems that chain speech recognition, a language model, and speech synthesis, one after the other, and which wait for the user to finish speaking before they begin to respond.

The output controls not only what is said but also how the speech is delivered and how the model behaves nonverbally, including emotional tone, stance, facial expression, and gesture. Because these decisions are made continuously and are conditioned on context, the model can take and yield the turn on the basis of what is being said rather than when the audio stops, so a pause for thought is not treated as the end of the turn. The same mechanism lets the model nod in agreement while the user is speaking, backchannel and confirm at natural points, and adjust the emotional content of its response to the user's speech and nonverbal behavior.

Additionally, Griffin perceives video as well as audio, where most interactive systems perceive audio alone. Visual input lets the model read the user's nonverbal communicative signals, such as gaze and facial expression, and it also gives the model access to the user's environment and to visual material the user chooses to share, such as their screen.

This is illustrated in the figure below.

Griffin Full Duplex Conversational Modeling

Emotionally Responsive Full Duplex Conversational Model

InputsSystem promptConversation historyUser AudioUser Video

Full Duplex MiniturnCCM

01 Perceive ...SpeechToneGestureFacial expressionScenescreen & background

02 ... and act accordingly

User

Model

Can you book a table for four at an Italian restaurant tomorrow night?

Of course, I’ll look for available reservations.

Silence

I am checking Italian restaurants with tables for four...

Mm-hm.Backchannel (no interruption)

Actually, make it sushi instead.⚡︎ Interrupted

Sure, I’ll switch to sushi options for tomorrow night.

Found a few — 7:00 pm and 8:30 pm are available.

Silence

PhasesBehaviors

User speaking

• Understand content and intent
• Track nonverbal cues (e.g., tone, gesture)
• Anticipate turn end

Model responding

• Generate speech, video and nonverbal behavior
• Keep listening for user signals
• Allow natural overlap

User backchannel

• Detect feedback (e.g., “mm-hm”, “I see”)
• Continue without interruption

Model continuing

• Incorporate backchannel
• Continue the response naturally
• Maintain engagement

User interrupting

• Detect interruption in real time
• Stop current response
• Adapt to new input

Model adapting

• Acknowledge and re-plan
• Generate updated response
• Maintain conversa­tional flow

Idle conversation

• Wait for next user input
• Stay engaged with appropriate nonverbal behavior

Across all phasesCall a toolChange demeanor depending on user emotion

OutputsTextControl signals for streaming audio and video generation

Perception continues during speech and silence.

The emitted outputs are converted to streaming audio outputs and together with streaming control signals are driving video generation as explained below.

Fast and Expressive Streaming Speech Generation

Griffin-Lite's speech generation produces high-quality speech at conversational latency and can clone a speaker's voice from about 10 seconds of audio. The model is a fast autoregressive diffusion transformer. It takes an encoding of the reference clip as a prefix and generates the new speech progressively, as increments of outputs and controls arrive from the conversational model, instead of waiting for the full utterance.

A compact continuous codec

Tavec v1.7

High fidelity. Extraordinary speed.

2,193×

real-time decoding

4.242

ViSQOL MOS-LQO

28.42M

decoder parameters

Continuous latents · Causal streamingDecoder throughput measured on H100.

Griffin's speech generation speed and fidelity start with how it represents sound. It relies on Tavec, a convolutional autoencoder that maps 48 kHz audio into a compact continuous latent: 40 values per frame, 100 frames per second, and no codebooks. Its fully causal (streaming-ready) decoder carries state across chunks, turning streamed latents into one seamless waveform, with audio packets as small as 10 ms.

This continuous representation supports straightforward regression-based flow matching and fully differentiable audio pipelines, without quantizer gradient approximations. Its compactness also makes long sequences manageable: a minute of speech corresponds to 6,000 latent frames, against 2.88 million samples of raw audio.

Griffin: Fast and Expressive Streaming Video generation

The other side of Griffin's audiovisual generation system is fast diffusion-based generation architecture. Previous Tavus models relied on extensive 3D priors, which limited what they could express: complex, human-like behavior such as hand and large body gestures was out of reach, and so were dynamic backgrounds.

We built it to meet three requirements:

1. Respond quickly to each incoming chunk of audio and expressive controls
2. Handle controls that change fast
3. Hold visual quality, lip sync, and identity consistency over long generations

Autoregressive latent video models typically compromise on at least one of these. The architecture we arrived at generates 720p video in real time, one latent at a time, and accepts streaming controls that let the conversation model direct nonverbal behavior such as gesturing, looking away, or changing emotion.

1DMD2

Few-step generation

Pretrained teacher

Many-stepBidirectional

DMD

Student

Few-stepBidirectional

DMD + adversarial loss

Few forward callsNot yet streaming

2Robust Teacher Forcing

Causal generation

Conditioning

ReferenceGround-truth history

Student

Few-stepCausal

Next latent

One latent at a timeStreaming enabled

3Self-Forcing

Long-horizon stability

Conditioning

ReferenceGenerated history

Student

Few-stepCausal

Next latent

Final step only

DMD

Gradient**Feed back / detached

Progressively longer rollouts

Robust Teacher Forcing

Reference

Ref

No compression

Oldest history

1…L

High compression

Intermediate

L+1…L+M

Medium compression

Most recent

L+M+1…L+M+S

Low compression

Target

Noise

No compression

DiT

Next latent

To build it, we distilled a large, bidirectional, many-step diffusion model into a few-step autoregressive generator. The generator takes a reference image together with streaming audio and controls, and produces one latent at a time. The distillation ran in three stages. First, we distilled the teacher into a few-step student with Distribution Matching Distillation. This student is fast but not autoregressive; it generates a long video as a single long chunk. Second, we converted it into an autoregressive model with teacher forcing, so as to arrive in an architecture that can generate in a few steps one latent at a time. Third, we trained with Self-Forcing so that we can run long autoregressive rollouts without drift, and we added a mechanism acting on the history frames, the previously generated frames the model continues from, to improve stability.

03

Griffin-Lite: Evaluations and benchmarks

Ahead of this preview, we ran Griffin-Lite through a set of evaluations at different levels, from a single component to a live conversation:

• The video generator on its own, against published streaming diffusion models, on latency, visual quality, and lip sync
• The full system on VideoFDB, NVIDIA's leading industry benchmark for full-duplex audio-visual conversation, which scores how a model reads a person's behavior and how it responds
• The full system in live one-minute video calls with people who did not know they were talking to a model

Video Generation

We compared the video generator of Griffin-Lite against four published streaming diffusion models in an audio-to-video setting, where each model is given speech and a reference image and has to produce the talking face.

Griffin-Lite produces one latent at a time and does not wait for future audio, so there is little delay between a piece of audio arriving and its effect appearing on screen. On H100s this averaged 0.43 seconds, half that of the next fastest method. In a conversation, this significantly cuts down the delay before the face reacts when the person interrupts or when a nod is due.

This is depicted in the diagram below, where the dark bar represents the time required by different models to generate the smallest chunk of video they can produce, and pink the time required to playback a video chunk. In the figure, we mark the best average and the worst case latency of each model.

Source audioGeneration timePlayback time

Griffin-Lite also scored highest on visual quality. It led every baseline on DOVER and FID, the two standard measures of video quality, and on THEval, a recently published framework built for talking heads. On lip sync it placed second on LSE-C, at 7.27. LSE-C rewards pronounced mouth movement, including movement past the point where it looks natural, which THEval accounts for.

THEval ↑

LiveTalk0.667

SoulX-FlashTalk0.696

AvatarForcing0.698

LiveAvatar0.657

Griffin Lite0.748

FID ↓

LiveTalk24.2

SoulX-FlashTalk21.4

AvatarForcing25.8

LiveAvatar20.9

Griffin Lite13.8

DOVER ↑

LiveTalk0.636

SoulX-FlashTalk0.583

AvatarForcing0.711

LiveAvatar0.702

Griffin Lite0.750

LSE-C ↑

LiveTalk7.05

SoulX-FlashTalk6.79

AvatarForcing7.73

LiveAvatar6.61

Griffin Lite7.27

Full-Duplex Conversation

NVIDIA's VideoFDB is a research and industry benchmark for evaluating the naturalness and accuracy of full-duplex audio-visual conversations. It examines how a model responds to and produces the signals of human interaction, including dialogue, gaze, facial expression, and body movement, measuring both conversation quality and response time. The benchmark scores a model on two tracks, generation and perception, each with its own rubric and its own leaderboard. NVIDIA conducts the evaluation independently using the published metrics and its own judge.

Griffin-Lite scored highest on both tracks against the benchmark's published baselines, which include commercial and open-source systems— and it is the only case where the same model is evaluated on both tracks.

The generation track tests whether the model produces the right behavior. Given a person's audio, it scores the model's speech and video together for fluency, affect matching, and whether the nonverbal behavior fits the moment, laughing with the person rather than after them. Griffin-Lite scored 3.83, over 1 full point ahead of the next-highest system and only 0.09 below the human reference at 3.92, which is more than 12x closer to human performance than any other system evaluated.

Generation

Cascaded A2AV — what the agent produced, audio and video together, in response to a dynamic event in the user's stream.

Overall ↑TOR ↑ / Timing ↓

Human ground truth3.9278% / 900 ms

Tavus Griffin Lite3.8362.8% / 1892 ms

Gemini 2.5 + Anam2.8044% / 2840 ms

Gemini 2.5 + Keyframe2.3931% / 3520 ms

VideoFDB's perception track tests the other side of the conversation, which is whether the model understands the moment. Given a person's audio and video, it scores the model's spoken response for fluency, conversational flow, and visual grounding, meaning whether the model uses what it sees, a pause with a glance away, for instance, rather than reacting to the words alone. Griffin-Lite scored 3.73, 0.29 points ahead of the strongest second best and 0.47 below the human reference at 4.20, marking Griffin-Lite as the most perceptive model on the benchmark among the 15 models evaluated, ahead of every frontier realtime model on the leaderboard.

Perception

AV2A — audio-visual in, audio out. Did the agent read the moment correctly? (A2A) rows are the audio-only variants of the same systems.

Overall ↑TOR ↑ / Timing ↓

Human reference4.2090% / 1400 ms

Tavus Griffin Lite3.7373.8% / 2232 ms

MiniCPM-o 4.5 (A2A)3.4472% / 920 ms

MiniCPM-o 4.53.4073% / 720 ms

Gemini 2.5 Flash Native3.1772% / 3160 ms

Gemini 2.5 Flash Native (A2A)3.1773% / 2760 ms

Gemini 3.1 Flash Live (A2A)3.0369% / 1240 ms

OpenAI gpt-realtime (A2A)2.9767% / 4440 ms

OpenAI gpt-realtime-mini (A2A)2.8869% / 5000 ms

Gemini 3.1 Flash Live2.8466% / 1720 ms

OpenAI gpt-realtime2.7572% / 5400 ms

OpenAI gpt-realtime-mini2.7366% / 5320 ms

VITA-1.5 (A2A)2.0061% / 800 ms

VITA-1.51.7658% / 400 ms

MiniOmni2 (A2A)1.7269% / 2760 ms

MiniOmni21.1964% / 3080 ms

VideoFDB also reports takeover-rate alignment, which measures how closely the model's decisions about when to speak match the timing in the reference conversations. Griffin-Lite scored 62.8% on generation and 73.8% on perception, the highest of any system on both tracks.

Face-to-face study

Griffin-Lite

48%

of participants (n=54) believed their partner was a real person.

Phoenix-4.5

2.4%

of participants (n=41) believed their partner was a real person.

Quantitative benchmarks are important, but some aspects of naturalness are difficult to measure. For every new model, we run a study with participants recruited through an independent research platform to understand how human-like it is in a real, live conversation. Participants have a short video call with the model without being told what it is, and afterward we ask whether they think their partner was a real person. Until now, almost no one has said yes.

On Phoenix-4.5, 2.4% of participants (n=41) believed their partner was a real person. On Griffin-Lite, nearly half did: 48% (n=54). This represents a milestone of, to the best of our knowledge, the first model to have ever passed the video Turing test.

Methodology

Participants were told they would be matched with another participant for a one-minute video call to discuss what they were looking forward to this year. Their partner was in fact a PAL powered by Griffin-Lite, generating her face, voice, and responses in real time. After the call, participants were asked to write down their partner's answer to the question and to rate their partner on several axes around naturalness and trustworthiness. They were also asked to rate the conversation itself: how well it flowed, and whether they felt their partner was really listening. Only at the end of the survey were participants asked whether it had crossed their mind that their partner might not be a real person, and if so, when. Every participant was then told that their partner had been an AI model.

Results

Of the 54 participants who spoke with Griffin-Lite, 26 believed their partner was a real person. Participants on both sides were confident in their answer: those who said real averaged 79% confidence, and those who said AI averaged 81%.

Over half of participants said the possibility had not crossed their mind during the call, and nearly all of that group went on to say their partner was real. Those who did suspect tended to suspect within the first 20 seconds.

Griffin-Lite was rated favorably on every axis. On a 7-point scale, participants on average rated it a 5.4 for seeming natural, 5.6 for seeming trustworthy, and 5.8 for whether they would enjoy talking with it again, a score that held at 5.4 among participants who said it was AI. The conversation scored 5.5 for whether participants felt their partner was really listening and 4.9 for flowing naturally, the lowest of the five.

04

Safety, limitations, and responsibility

Given Griffin-Lite is the first model to pass the Turing test, we understand the responsibility to be thoughtful on the risks and release strategies.

The same properties that make Human Interaction Models powerful interfaces for natural communications between human and machine allow them to deceive a human into believing it is not AI.

We believe further alignment and safety procedures are required for safe release, and we are working on safe disclosure features, as well as with organizations tackling AI safety. Contact us if you'd like to participate in these evaluations and testing.

Griffin-Lite will not be available for use for customers at this time, though it is available for select trusted testers as a research preview. To request access, you can submit this form. We anticipate releasing Griffin very soon after these safety concerns are addressed.

05

What's next

Griffin is an early step toward computers that people can work with instead of operate. As these models improve, they can bring knowledge and emotional understanding together and play a larger part in how people learn, work, and get help.

For a student, that could mean a tutor who notices when an explanation isn't making sense and tries another one. For someone at work, it could mean rehearsing a hard conversation with a counterpart who reacts the way a real person might. For a customer, it could mean holding a broken part up to the camera and working out the fix together, without knowing what the part is called.

In each of those cases, the person doesn't have to turn what they need into the right command or menu first. They can explain it the way they would to another person, and the model can work out what they mean from their words, their face, and the moment. That's what we mean by human computing, and Griffin is our biggest step yet toward it.

Credits

Griffin was built by the Tavus research and engineering teams. Our thanks to Baseten, Daily and Cerebrium for the infrastructure behind the preview, to Queen Mary University of London for research collaboration, and to NVIDIA for building and scoring the Video Full-Duplex Benchmark.

Citation

Tavus Research, “Griffin: The First Human Interaction Model”, Tavus, October 2026.

Build with Tavus today

Griffin isn’t on the Tavus platform yet. It’ll come once we’ve worked out how to release it safely. Until then, you can build PALs, AI you talk to face to face, on the models 150,000 developers and businesses already use: Phoenix generates the face, Raven sees and understands you, and Sparrow knows when to talk.