Skip to content
Model infoModelsSnapshot: 15 Oct 2026

Tavus / Tavec

Tavec v1.7

Continuous neural audio codec used within Griffin, with a causal streaming decoder and no codebooks.

Audio processing
restrictedClosed weightsAudio processingAudio codec
Official documentation (opens in a new tab)Reviewed 14 Oct 2026
6 of 17 specification fields collected
Audio sample rates48 kHzSee specification notes below
LanguagesNot collectedAwaiting source review
Input billingNot collectedTavus research
Output billingNot collectedNo inferred price

Specifications

Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.

Context / input limit
Not collected

Exact model-version information has not been established; unknown does not mean unsupported.

Maximum output
Not collected

Exact model-version information has not been established; unknown does not mean unsupported.

Context caching
Not collected

Exact model-version information has not been established; unknown does not mean unsupported.

Reasoning
Not collected

Exact model-version information has not been established; unknown does not mean unsupported.

Tool calling
Not collected

Exact model-version information has not been established; unknown does not mean unsupported.

Structured output
Not collected

Exact model-version information has not been established; unknown does not mean unsupported.

Streaming
SupportedSource for Streaming (opens in a new tab)
Batch processing
Not collected

Exact model-version information has not been established; unknown does not mean unsupported.

Fine-tuning
Not collected

Exact model-version information has not been established; unknown does not mean unsupported.

Parameters
28.42 million decoder parametersSource for Parameters (opens in a new tab)

Decoder only; the total encoder-plus-decoder model size is not established.

Architecture
Convolutional autoencoder with continuous latents and a fully causal streaming decoder; no codebooksSource for Architecture (opens in a new tab)
Knowledge cutoff
Not collected

Exact model-version information has not been established; unknown does not mean unsupported.

Languages
Not collected

Exact model-version information has not been established; unknown does not mean unsupported.

Release date
Not collected

Exact model-version information has not been established; unknown does not mean unsupported.

Input modalities
AudioSource for Input modalities (opens in a new tab)
Output modalities
Continuous audio latents, AudioSource for Output modalities (opens in a new tab)

Autoencoder representation and waveform reconstruction, not text-to-speech or speech recognition.

Audio sample rates
48 kHzSource for Audio sample rates (opens in a new tab)

Access & pricing

Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.

Tavus research

Provider source (opens in a new tab)
API model ID
Not collected

No currently selectable exact public API identifier verified.

Access
Restricted research only; no generally available public model endpoint or standalone price verified.
Regions
Privacy policy describes US processing and possible transfers to service providers in other countries. Exact model-serving location is not established.
latentRepresentation
40 values per frame at 100 frames per second; 6,000 latent frames per minute of speech.Source for latentRepresentation (opens in a new tab)
streamingPackets
Audio packets as small as 10 ms with decoder state preserved across chunks.Source for streamingPackets (opens in a new tab)
accessRestriction
Embedded in Griffin research; no independent serving or download endpoint verified.Source for accessRestriction (opens in a new tab)

Endpoint pricing has not been collected. Check the provider for applicable rates.

Data & enterprise terms

Default privacy terms permit service improvement and AI research/training. Biometric retention is limited by the stated purpose or one year after last interaction, whichever comes first. Enterprise zero-retention arrangements are separate; no universal zero-retention promise is inferred.

Read the service policy (opens in a new tab)

Caching

Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.

01 / INITIAL INPUTProcess & retain

The system processes eligible input and retains reusable state.

02 / LATER REQUESTCheck available state

Reuse depends on matching input, session state and the cache policy.

03 / CACHE HITReuse & generate

Reusing state can reduce repeated work. Any charges follow the billing rules below.

Tavus research

Not verified

Internal streaming state, saved faces and conversation memories do not establish billable prompt caching, cache lifetime or discounts.

How it works
Not collected

Internal streaming state, saved faces and conversation memories do not establish billable prompt caching, cache lifetime or discounts.

Minimum reusable input
Not collected

Internal streaming state, saved faces and conversation memories do not establish billable prompt caching, cache lifetime or discounts.

Lifetime & refresh
Not collected

Internal streaming state, saved faces and conversation memories do not establish billable prompt caching, cache lifetime or discounts.

How to enable it
Not collected

Internal streaming state, saved faces and conversation memories do not establish billable prompt caching, cache lifetime or discounts.

Usage fields
Not collected

Internal streaming state, saved faces and conversation memories do not establish billable prompt caching, cache lifetime or discounts.

Billing rules
Not collected

Internal streaming state, saved faces and conversation memories do not establish billable prompt caching, cache lifetime or discounts.

Limits & data boundaries
Not collected

Internal streaming state, saved faces and conversation memories do not establish billable prompt caching, cache lifetime or discounts.

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Caching reviewed 14 Oct 2026

What the measurements say

Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.

The record behind the model

Why it is included

Named model developed by Tavus in its proprietary hosted/research stack. Reviewed primary documentation provides no downloadable weights for this model; service terms restrict source derivation and independent avatar use. Open SDKs and third-party models in the surrounding pipeline are separate.

Classification evidence (opens in a new tab)

Lifecycle dates

Named in Griffin research report; first release date unknown1 Oct 2026Source (opens in a new tab)

Sources reviewed