AI BriefingGemini 3.5 Transcribe Now Available on AI Gateway
Key point
Google's Gemini 3.5 Transcribe model has been added to Vercel AI Gateway, supporting real-time and full-audio transcription.
Details
Google's Gemini 3.5 Transcribe model is now officially supported on Vercel's AI Gateway. The model converts audio input to text and is available in two variants depending on the usage method.
google/gemini-3.5-transcribe: Transcribes entire recorded audio files in a single request.google/gemini-3.5-transcribe-live: Streams audio via WebSocket, updating transcription results in real-time as recording progresses.
AI SDK V7 introduces new streaming transcription capabilities, allowing audio streams read directly from the microphone to be passed to the model via streamTranscribe. Additionally, previously stored audio files can be converted to text in a single request using the transcribe function.
AI Gateway provides a unified API for model invocation, usage and cost tracking, failover, and performance optimization, applying provider pricing directly with no platform fees or markups. BYOK (Bring Your Own Key) requests also incur no platform fees on inference costs.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Fake Customer Service Pages Induce ChatGPT and Gemini to Display Scam Numbers