Skip to content

Amazon Nova 2.5 Sonic

Coming soon

Amazon serves this model through a bidirectional streaming API. AnyRouter does not route speech-to-speech requests yet.

amazon/nova-2.5-sonic

Speech · Oct 2026 · 0 upstreams

Amazon Nova 2.5 Sonic is a speech-to-speech model for real-time voice conversations. Amazon reports lower latency and better tool calling than Nova 2 Sonic.

Coming soon

Announced, not routable yet. Follow the changelog for the launch.

context
—
announced price
TBA
Speech output
Voice — voice

This provider does not publish an enumerable voice list.

PriceNot published
ArchitectureTransformer
Categoryaudio
ReleasedOct 5, 2026
Modalities
→
Capabilities
Speech models return raw audio bytes, not JSON — write the response body to a file. They cannot be called through chat completions.