Microsoft Developing AI Processor 'Athena'
Reportedly, Microsoft's aim is to avoid costs associated with using Nvidia's GPUs.
A report by The Information citing unnamed sources suggests that Microsoft has been developing a chip that will support large language models (LLMs).
The chip, code-named "Athena," reportedly has been in the works since at least 2019, and is currently being tested by "a small group of Microsoft and Open AI employees."
A corresponding report by The Verge suggests that Micorosft's aim is to avoid costs associated with using Nvidia's GPUs. Other cloud service providers make their own AI chips, and so Microsoft is doing the same, it's said.
A story by Thurrott.com indicated that more than 300 people are working on Athena, with an AI chipset expected next year. It's not clear if this AI chip will be part of Azure services.
"It's likewise not clear what this means to Microsoft’s supercomputer partnership with Nvidia," the Thurrott.com story added.
I requested further information from Microsoft, but there's been no word back so far. Update 4/18: "Microsoft has nothing to share."
Kurt Mackie is senior news producer for 1105 Media's Converge360 group.