Microsoft’s latest AI chip goes head-to-head with Amazon and Google
ThinkingNews Desk · how this was written
Microsoft has launched the Maia 200, its second-generation AI inference accelerator built on TSMC’s 3 nm process and succeeding the 2023 Maia 100. The chip, containing over 100 billion transistors, is being deployed today in Azure’s US Central data-center region and is designed to run AI models faster and more efficiently, delivering roughly three times the FP4 performance of Amazon’s Trainium and surpassing Google’s TPU in FP8 benchmarks.
Written from all 3 reports below, not from any single one.
How it was reported
- TechCrunch·Microsoft announces powerful new chip for AI inference
Microsoft has launched the Maia 200 chip, designed for AI inference. The chip is technically outfitted to run AI models at faster speeds and with more efficiency. It follows the Maia 100, released in 2023.
- The Verge·Microsoft’s latest AI chip goes head-to-head with Amazon and Google
Microsoft's Maia 200 AI chip delivers 3 times the FP4 performance of Amazon's Trainium. It has over 100 billion transistors, handling large-scale AI workloads. The chip outperforms Google's TPU in FP8 performance.
- TechMeme·Microsoft unveils the Maia 200, its 2nd-generation AI accelerator built on TSMC's 3nm process, deploying today in its Azure US Central data center region (Tom Warren/The Verge)
Microsoft unveils the Maia 200, its 2nd-generation AI accelerator. The chip is built on TSMC's 3nm process and starts deploying in Azure's US Central data center region. It succeeds Microsoft's first in-house AI chip.
Related stories
- Microsoft releases ASSERT, an open-source framework that lets developers generate and run AI behavior tests using natural-language descriptions (Ram Iyer/TechCrunch)5 outlets
- Build for the new AI era with Microsoft and NVIDIA3 outlets
- Microsoft used agentic AI to make its quantum chip 1,000 times more reliable, then cut its timeline to a scalable quantum computer in half3 outlets
- Nvidia and AI chip startup d-Matrix are combining their hardware in a new system to power AI models, the latest example of Nvidia partnering with rivals (Phoebe Liu/The Information)3 outlets