Microsoft on Monday unveiled the second generation of its in-house artificial intelligence chip, the Maia 200, along with software tools aimed at rivaling Nvidia’s key advantages with developers.
The new chip will go live this week in a data center in Iowa, with plans for a second facility in Arizona. Microsoft first introduced the Maia chip in 2023, and the Maia 200 marks the company’s next step in competing in the AI hardware space.
Alongside the chip, Microsoft is offering a suite of software tools, including Triton, an open-source programming tool developed with contributions from OpenAI. Triton is designed to perform similar tasks to Cuda, Nvidia’s proprietary software widely regarded as a major competitive edge.
The Maia 200 uses Taiwan Semiconductor Manufacturing Co.’s 3-nanometer process and high-bandwidth memory chips, although slightly older than those in Nvidia’s upcoming Vera Rubin chips. Microsoft has also added a significant amount of SRAM, a type of memory that speeds up AI systems when handling multiple user requests simultaneously.
Industry analysts note that major cloud providers, including Microsoft, Google, and Amazon Web Services, are increasingly producing their own AI chips, reducing dependence on Nvidia. Companies like Cerebras Systems and Groq also use SRAM technology to boost performance for AI applications.
The release of Maia 200 highlights the intensifying competition in AI hardware, as cloud providers and chipmakers race to deliver faster, more efficient systems for the growing demand in artificial intelligence.
