Gemini 3.5 Transcribe now available on AI Gateway
By Steven Van ·
Gemini 3.5 Transcribe from Google is now available on AI Gateway for recorded and live audio: google/gemini-3.5-transcribe transcribes a complete audio…
Google's Gemini 3.5 Transcribe is now available on Vercel's AI Gateway, in two versions: google/gemini-3.5-transcribe for transcribing a complete audio file in one request, and google/gemini-3.5-transcribe-live for streaming transcription over a WebSocket. Both automatically detect more than 85 languages, including mid-conversation switches, and accept custom vocabulary to improve recognition of names, technical terms, and uncommon spellings. The complete-recording model can also identify speakers and return word-level timestamps.
Streaming transcription requires AI SDK 7 and the latest AI Gateway provider. It can be tried in the model playground, alongside Vercel's other transcription models, as described in the announcement.
