Microsoft unveils Maia 300, to be deployed in 2027.
Microsoft unveiled Maia 300, a custom AI inference accelerator scheduled for deployment in 2027.
OriginalSourceAxiosUnited StatesLeanCenterCredibilityPublished09:00 UTCRatings: AllSides, MBFCSourceAxiosUnited StatesLeanCenterCredibilityPublished09:00 UTCRatings: AllSides, MBFC
“Every major Nvidia customer is now a competitor as Microsoft reveals new Maia chip”
Microsoft unveils Maia 300, to be deployed in 2027.
Key points
- Microsoft unveiled Maia 300, its custom AI accelerator for inference.
- Maia 300 aims to deliver 40% more AI tokens per megawatt than GPUs.
- Deployment will begin in 2027, becoming primary inference system for Microsoft.
- The move signals Nvidia customers seeking cost, performance, and control gains.
- Industry players like Google, OpenAI, Amazon, and Anthropic explore own chips.
How it got here
Axios publishes the article.
concise reads it.
The Brief ranks it 8th.
8 8 Microsoft unveils Maia 300, to be deployed in 2027.
Microsoft unveiled Maia 300, a custom AI inference accelerator scheduled for deployment in 2027.