Microsoft AI models are now available on AI Gateway
By Steven Van ·
Microsoft's MAI-Voice-2.1, MAI-Voice-2.1-Flash and MAI-Transcribe-2 Streaming can be called through AI Gateway, billed at listed rates with no markup.
Three Microsoft AI (MAI) audio models are now available on Vercel's AI Gateway. Vercel says AI Gateway is one of a limited number of platforms offering access to MAI models, and that MAI's focus on security and zero data retention (ZDR) fits its aim of giving users control over their data and transparency on training use. The announcement lists the models:
- MAI-Voice-2.1 (microsoft/mai-voice-2.1) generates expressive speech in 23 languages and keeps a consistent speaker across longer passages. Vercel suggests it for narration, audiobooks, podcasts and lessons.
- MAI-Voice-2.1-Flash (microsoft/mai-voice-2.1-flash) offers the same multilingual speech generation with lower latency, aimed at voice agents, assistants and spoken replies.
- MAI-Transcribe-2 Streaming (microsoft/mai-transcribe-2-streaming) returns partial transcripts while audio is still arriving, so applications can show live captions or follow a conversation without waiting for the full recording.
AI Gateway bills the models at their listed rates, with no platform fee or markup on inference. The speech models work through generateSpeech in AI SDK 7, and the transcription model takes a stream of audio chunks.
Vercel
The platform for frontend developers — deploy, preview, and scale web apps and AI agents with zero config.
View Vercel →