Models / Deepgram
Transcribe

Deepgram Nova-3

Real-time speech-to-text for voice agents

Try now
Read Docs

This model is not available on Together’s Serverless API.

Related models
View All
No items found.
View All
Model specifications
Model data
  • Model provider
    White stylized letter D on a black circular background.
    Deepgram
  • Type
    Transcribe
  • Price

    $0.0077 / min + GPU hourly

  • Input modalities
    Audio
  • Output modalities
    Text
  • Released
    February 11, 2025
  • Category
    Transcribe
Quickstart docs
Deploy Model

Launching soon

Want to be notified when the model is available on Together AI?

You're on the list

We'll email you when the endpoint goes live. Until then, 200+ models are ready to call.

Browse Models
Continue