AMD Unveils Powerful AI Chips to Challenge Nvidia, Attracts Interest from Meta and Microsoft
AMD has released new accelerators and processors, including the Instinct MI300X accelerator and the Instinct M1300A accelerated processing unit (APU), designed for running large language models (LLMs) and powering faster AI training. The MI300X accelerator offers 1.5 times more memory capacity than its predecessor and is comparable to Nvidia's H100 chips in training LLMs. AMD has partnered with Microsoft to deploy the MI300X in Azure virtual machines, and Meta will use MI300 processors in its data centers. Additionally, AMD introduced the Ryzen 8040 processor series, which integrates neural processing units (NPUs) and offers improved AI processing performance. The company expects manufacturers like Acer, Asus, Dell, HP, Lenovo, and Razer to release products with Ryzen 8040 chips in Q1 2024.