- Microsoft unveils Maia 200 AI hardware
- Maia 200 reportedly offers more performance and efficiency than AWS and GCP rivals
- Microsoft will use it to help improve Copilot internally, but it will also be available to customers
Microsoft has unveiled Maia 200, its “next big milestone” in supporting the next generation of AI and inference technology.
The company’s new hardware, the successor to the Maia 100, will “dramatically change the economics of large-scale AI,” offering a significant upgrade in performance and efficiency as it claims the market.
The launch will also look to push Microsoft Azure as a great place to run AI models faster and more efficiently, as it looks to take on its big rivals Amazon Web Services and Google Cloud.
Microsoft Maia 200
Microsoft says the Maia 200 contains over 100 billion transistors built on the TSMC 3nm process with built-in FP8/FP4 tensor cores, a redesigned memory system with 216GB HBM3e at 7TB/s and 272MB on-chip SRAM.
All this contributes to the ability to deliver over 10 PFLOPS of 4-bit precision (FP4) and around 5 PFLOPS of 8-bit (FP8) performance – easily enough to run even the largest AI models today, and with room to grow as the technology evolves.
Microsoft says Maia 200 delivers 3x the FP4 performance of third-generation Amazon Trainium hardware and FP8 performance over Google’s seventh-generation TPU, making it the company’s most powerful inference system to date.
And thanks to its optimized design, which sees the memory subsystem centered on narrow-precision data types, a specialized DMA engine, on-die SRAM and a specialized NoC fabric for high-bandwidth data movement, the Maia 200 is able to keep more of a model’s weight and data locally, meaning fewer devices are required to run a model.
Microsoft is already using the new hardware to power its AI workloads in Microsoft Foundry and Microsoft 365 Copilot, with wider customer availability coming soon.
It is also rolling out Maia 200 to its US Central data center region now, with additional deployments on the way to its US West 3 data center region near Phoenix, Arizona, and additional regions to follow soon.
For those who want an early look, Microsoft is inviting academics, developers, frontier AI labs, and contributors to open source model projects to sign up for a preview of the new Maia 200 software development kit (SDK) now.
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews and opinions in your feeds. Be sure to click the Follow button!
And of course you can too follow TechRadar on TikTok for news, reviews, video unboxings, and get regular updates from us on WhatsApp also.
