Lower latency, lower cost, more possibilities

We’re excited to introduce major improvements to our API’s inference latency, with the majority of audio files now completing in well under 45 seconds regardless of audio duration.

announcements api audio cost files improvements inference latency major