TechFlow news, on June 14, U.S. semiconductor company AMD released new details on June 13 about its artificial intelligence (AI) chips that will challenge market leader Nvidia Corporation.
AMD stated that its most advanced AI graphics processing unit (GPU), the MI300X, will begin volume production in the third quarter and mass production in the fourth quarter. The chip and CDNA architecture are specifically designed to meet the demands of large language models and advanced AI models. Compared to other chips, the MI300X offers greater memory capacity—supporting up to 192GB of memory.
AMD also announced Infinity Architecture, which integrates eight MI300X accelerators into a single system, enabling AI applications to combine eight or more GPUs.
AMD CEO Lisa Su said in a speech that artificial intelligence represents the company’s “most important and strategically significant long-term growth opportunity.”
If developers and server manufacturers adopt AMD’s AI chips as an alternative to Nvidia products, this could open up a major untapped market for the chipmaker.
This move could exert downward pricing pressure on Nvidia's GPUs, including models such as the H100. Lowering GPU prices could help reduce the overall costs associated with running resource-intensive generative AI applications.
