AI Briefing#microsoft-ai
The latest AI and developer news about #microsoft-ai, with the original source and a short summary.
The latest AI and developer news about #microsoft-ai, with the original source and a short summary.
The ~50-person research group is ending operations after being reorganized into Microsoft AI, which prioritizes scaling over fundamental research.
Trained on 16,000 GB200 GPUs for 38 days with 50T tokens, claiming 4–12x compute efficiency improvement over DeepSeek-V4-Pro-Base
The integration includes MAI-Voice-2.1, MAI-Voice-2.1-Flash, and MAI-Transcribe-2 Streaming with no platform fee or markup on inference.
MAI-Image-2.6 secured 2nd place on the Arena text-to-image leaderboard, surpassing models from Google, Meta, and xAI.
Microsoft AI introduced MAI-Code-1.1-Flash to GitHub Copilot, improving performance and speed while reducing costs to a quarter.
Microsoft has unveiled a new MAI model lineup including a high-performance image generation model and a high-speed voice model.
Microsoft AI has released 7 new MAI models and unveiled Frontier Tuning technology optimized for user workflows.
Microsoft has announced 3 new high-performance MAI models available on Foundry.