Microsoft Launches Maia 200 AI Accelerator Chip to Compete with Amazon and Google
By
Tom Warren
Warm and crisp on the edges. A bagel with a bit of bite.
Summary
Microsoft announces the Maia 200, its latest in-house AI accelerator chip built on TSMC's 3nm process. The chip features over 100 billion transistors and is designed to handle large-scale AI workloads. Microsoft claims the Maia 200 delivers 3 times the FP4 performance of Amazon's third-generation Trainium chip and outperforms Google's seventh-generation TPU in FP8 performance. The chip is positioned to compete directly with AI chips from Amazon and Google, with Microsoft stating it can run today's largest AI models and has headroom for future, even larger models.
Key quotes
· 4 pulledMicrosoft says its Maia 200 AI accelerator 'delivers 3 times the FP4 performance of the third generation Amazon Trainium, and FP8 performance above Google's seventh generation TPU.'
'Maia 200 can effortlessly run today's largest models, with plenty of headroom for even bigger models in the future,' says Scott Guthrie, executive vice president of Microsoft's Cloud.
Each Maia 200 chip has more than 100 billion transistors, which are all designed to handle large-scale AI workloads.
Built on TSMC's 3nm process, Microsoft says its Maia 200 AI accelerator...
You might also wanna read
Apple Announces M5 Chip with Major AI Performance Improvements
Apple has announced its new M5 chip, representing a significant advancement in AI performance and overall chip capabilities. Built using thi
Google Announces Eighth Generation TPUs with Two Specialized Chip Architectures for AI Training and Serving
Google has announced its eighth generation Tensor Processing Units (TPUs) at Google Cloud Next, featuring two distinct chip architectures de
blog.google·1mo ago
AWS Announces Trainium3 AI Chip and Reveals Trainium4 Roadmap with Nvidia Compatibility
AWS announced its new Trainium3 AI training chip at re:Invent 2025, featuring 3nm technology and impressive specifications. The company also
techcrunch.com·6mo ago
Google Unveils Ironwood: Seventh-Generation Tensor Processing Unit with 4.7x Performance per Watt Improvement
Google has announced Ironwood, its seventh-generation Tensor Processing Unit (TPU), which represents a significant advancement in AI hardwar
blog.google·6mo ago
TSMC to Manufacture Advanced 3-Nanometer AI Chips in Japan
TSMC, the world's largest contract chipmaker, announced it will manufacture advanced 3-nanometer semiconductors in Japan to meet booming AI
apnews.com·3mo ago
AMD Unveils MI355X GPU and Future Roadmap for AI Advancements
AMD announces new MI355X GPU with improved AI performance and efficiency, software enhancements, rack-scale solutions, and future roadmap fo
morethanmoore.substack.com·11mo ago