Gemini 3.8 Live models now available on AI Gateway
Source
vercel.com
Author
Kevin Dawkins
Date
Context
Vercel has added Google's Gemini 3.8 Live models to AI Gateway for applications that hold spoken conversations. Its integration example uses short-lived access credentials to open a WebSocket connection, then a model adapter to translate outgoing and incoming realtime events.
That gives developers a Gateway entry point for a continuing voice session, including the Extended Thinking variant described in the model announcement. The example also handles transcript updates, completion, errors and connection closure, showing the session lifecycle an application must manage.
Terms in this piece · Glossary
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
