On June 13, Advanced Micro Devices (AMD) revealed new details about an artificial intelligence (AI) chip that could challenge market leader Nvidia.
California-based AMD said volume production of its most advanced AI graphics processing unit (GPU), the M1300X, will begin in the third quarter of 2023, with mass production beginning in the fourth quarter. AMD's announcement is the biggest challenge to Nvidia , which currently dominates the AI chip market with more than 80 percent market share. GPUs are the chips that companies like OpenAI use to build cutting-edge artificial intelligence programs like ChatGPT. They feature parallel processing capabilities and are optimized for processing large amounts of data simultaneously, making them ideal for tasks that require high-speed, efficient graphics processing.
AMD announced its latest MI300X chip and CDNA architecture specifically developed to meet the demands of large language and advanced AI models. With a maximum memory capacity of 192 GB, the M1300X can hold larger AI models than other chips l ike Nvidia's H100 chip, which supports up to 120 GB of memory. AMD's Infinity Architecture technology combines eight M1300X accelerators into a single system, similar to similar systems from Nvidia and Google that integrate eight or more GPUs for AI applications.
Speaking to investors and analysts in San Francisco, AMD CEO Lisa Su emphasized that artificial intelligence represents the company's "most important and strategically important long-term growth opportunity."
“We think the data center AI accelerator [market] will grow from around $30 billion this year to over $150 billion in 2027, a compound annual growth rate of over 50 percent.”
If developers and server makers adopt AMD's "accelerator" AI chips as an alternative to Nvidia's offerings, it could open up a significant untapped market for chipmakers. AMD, best known for its traditional computer processors, would benefit from a potential shi ft in demand. While AMD didn't reveal specific pricing details, the move could put downward pressure on the prices of Nvidia's GPUs, including models like the H100, which can cost as much as $30,000 or more. Lower GPU prices have the potential to help reduce the overall costs associated with running resource-intensive generative AI applications.
