Microsoft unveils Maia 300 AI chip
Microsoft on Monday unveiled the Maia 300, its third-generation AI chip. According to Microsoft, the new chip will be optimized to handle the latest reasoning AI models and agents, which should result in lower operating costs. “We expect Maia 300 to deliver 40% more tokens per megawatt compared to leading GPU-based systems,” said Scott Guthrie, Microsoft’s executive vice president for cloud and…
On Monday, Microsoft unveiled its latest AI chip, the Maia 300, marking the third generation in the series. The company claims that this new chip is optimized to efficiently manage the most recent reasoning AI models and agents, which will likely lead to reduced operating costs. Scott Guthrie, Microsoft’s executive vice president for cloud and AI, stated in a blog post that Maia 300 is expected to generate 40% more tokens per megawatt in comparison to leading GPU-based systems.
The Maia 300 builds upon the foundation laid by its predecessor, the Maia 200, which was introduced in January. Today, the Maia 200 supports Microsoft Copilot workloads, with GitHub Copilot and Microsoft Foundry set to follow by the end of 2026, according to the company.
Microsoft intends to release the Maia 300 after the turn of the year. This shift to a homegrown AI chip will reduce Microsoft's reliance on those from Nvidia and AMD, as the company moves more workloads to its own AI chips, according to The Verge. Scott Guthrie, Microsoft’s executive vice president for cloud and AI, added that they plan to scale the Azure Maia family to multi-gigawatt deployments beginning in 2027, making Maia the primary inference system for Microsoft products and services.
Written by urgent.news from Computerworld's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
- Microsoft introduces Maia 300 for large-scale AI deployment thearabianpost.com
