AWS's push into custom silicon via its Graviton4 CPUs and Trainium series accelerators is set to erode NVIDIA's market share in the AI chip sector.

Trainium 2 Chip Starting at the silicon level, new AWS CEO Matt Garman kicked off his keynote by highlighting the company's substantial investments in custom chips over the last decade.

Amazon Web Services hints at 1000 Watt next-gen Trainium AI chip — AWS lays the groundwork for liquid-cooled data centers to house new AI chips. Story by Jeff Butts • 11mo.

Amazon could spark a rivalry with Intel and AMD by updating its Graviton4 chip with 600 Gbps bandwidth. AWS also leverages AI ...

AWS is already Anthropic’s primary cloud provider, and the OpenAI rival will now also primarily use Trainium and Inferentia chips to train and deploy its foundation models.

For Trainium3, AWS expects another 4x performance gain for its UltraServers, for example, and it promises to deliver this next iteration, built on a 3-nanometer process, in late 2025.

Kenny Huynh was in charge of custom die package, card, server, rack mechanical, and thermals for AWS' Trainium 1 and 2 servers. The Amazon EC2 Trn2 Instance server sled – AWS. During his ...

AWS announced plans for an AI supercomputer, UltraCluster, with Trainium 2 chips at re:Invent. AWS may be able to reduce reliance on Nvidia by developing its own AI infrastructure. Apple said it's ...

AWS has been building its own custom processors since it acquired Annapurna Labs back in 2015 for about $350 million. Its first chip, Graviton, was an ARM-based design that easily slid into its ...

AWS announced plans for an AI supercomputer, UltraCluster, with Trainium 2 chips at re:Invent. AWS may be able to reduce reliance on Nvidia by developing its own AI infrastructure. Apple said it's ...

During his five-and-a-half-year stint at the cloud giant, Huynh also worked on AWS' custom Nvidia platforms for H100, GH200, and GB200. He has a number of patents, including for processor heat ...