Microsoft will launch a preview of its custom Azure Cobalt 100 chips at its Build developer conference next week.

In comments reported by TechCrunch, Scott Guthrie, EVP of Microsoft’s Cloud and AI group, told analysts that the chips, which were first announced in November 2023, will offer 40 percent better performance over other Arm chips on the market.

Guthrie said that Snowflake and Adobe were among customers who were already using the new chips.

The Microsoft Azure Cobalt CPU is designed for general workloads, with a focus on performance per watt. When the chips were first announced, Microsoft said they were being used for internal Microsoft products such as Azure SQL servers and Microsoft Teams.

The Cobalt 100 has 128 Neoverse N2 cores on Armv9 and 12 channels of DDR5, and is based on Arm’s Neoverse Genesis CSS (Compute Subsystem) Platform.

Microsoft also told analysts that it is planning to make AMD’s MI300X accelerators available to customers from next week. The GPUs will be sold through the company’s Azure cloud computing service.

First announced in December 2023, the MI300X accelerator is also built on CDNA 3 architecture and has 1.5× more memory capacity (192GB) and 1.7X more peak theoretical memory bandwidth (5.3TBps) than the previous M1250X version, delivering nearly 40 percent more compute units.

AMD also claims its new MI300X GPUs exceed the speed of Nvidia's H100 chips, offering 1.3 petaflops of FP16 and 2.6 petaflops of FP8 performance.

According to TechCrunch, Guthrie described the MI300X as the “most cost-effective GPU out there right now for Azure OpenAI.”

In addition to making the MI300X available, Microsoft has also developed its own AI accelerator in-house called the Azure Maia 100. Built on Arm architecture, the chip has been optimized for artificial intelligence tasks and generative AI and is used in the company’s cloud data centers.