AMD (AMD) debuted its latest artificial intelligence chips during its Advancing AI 2024 event in San Francisco, Calif., on Thursday.

The flurry of announcements comes as AMD continues to battle AI market leader Nvidia (NVDA) and looks to take greater market share from longtime rival Intel (INTC) in the server CPU space.

The company showed off its new 5th Gen AMD EPYC central processing units (CPU) for servers, provided details on its Instinct MI325X AI accelerator, and showed off its Ryzen AI PRO 300 processors for AI PCs for enterprise customers.

Shares of AMD fell more than 4% in afternoon trading.

AMD contended that its top-of-the-line 5th Gen EPYC 9965 chips, which cost $14,813, beat out Intel’s fifth-generation Xeon server chips — with servers running AMD’s processors offering 4x faster video transcoding times; 3.9x improvements in time to insights, the time it takes to turn data into useable information for science and high-performance computing applications; and 1.6x performance per core in virtualized infrastructure.

In other words: AMD wants you to know its chip can outpace Intel’s in certain scenarios.

Intel debuted its next-generation Xeon 6 chip in September, but AMD says it hasn’t been able to get its hands on the processor to test it against its EPYC chip.

On the AI accelerator side, AMD offered more details about its MI325X chip. The company says the data center processor beasts out Nvidia’s popular H200 AI chip when it comes to memory bandwidth and capacity, with the MI325X providing 256GB of HBM3E, a form of high-bandwidth memory used in AI processors.

AMD said the MI325X features 1.8x higher memory capacity than Nvidia’s H200 and 1.3x more bandwidth. According to the chipmaker, companies including Dell (DELL), Eviden, Gigabyte, HPE (HPE), Lenovo (LNVGF), and Super Micro Computer (SMCI) will begin offering MI325-based platforms in the first quarter of 2025.

The company also said it’s gearing up for its MI350X, the follow-up to the MI325X, which will debut in the second half of next year. Nvidia, meanwhile, is working with customers to deploy its next-generation Blackwell-based AI servers.

The data center has become the new battlefield for AMD, Nvidia, and Intel, as the companies seek to take advantage of the ongoing AI gold rush and grab as many customers as possible.

And that’s translated to some big time cash for AMD and Nvidia. In its most recent quarter, AMD reported record Data Center sales of $2.8 billion, up 115% year over year. Still, that’s a far cry from Nvidia’s Data Center business, which reported revenue of $26.3 billion, jumping 154% year over year.