Intel debuts super-fast AI chips – looking to match AMD in the race for AI chip supremacy
Xeon 6 and Gaudi 3 chips are designed for high-performance AI workloads
When you purchase through links on our site, we may earn an affiliate commission.Here’s how it works.
Intelhas been having a challenging time in recent years, particularly following the launch of theiPhonewithApple silicon, and the rise of an AI and data center chip industry largely dominated by Nvidia and AMD.
The company isn’t taking these developments lying down, though, and hasannouncedthe release of its Xeon 6 chips and new Gaudi 3 AI accelerators.
The Intel Xeon 6900 P-core series, as the chips are known, offers up to 128 cores for extremely intense AI workloads, doubling the performance of its predecessors, with higher core counts, more memory bandwidth, and embedded AI acceleration.
Power boost
On the Gaudi side, the new Gaudi 3 AI Accelerator chips, aimed atgenerative AI, come with 64 Tensor CPU cores, eight matrix multiplication engines, and 128GB of HBM2e memory, offering up to 20% more throughput and twice the price and performance overNvidia’s H100 chips for LLaMa 2 70B inference.
The specs Intel is touting for the Gaudi 3 and Xeon 6 chips are really impressive, and would signal the company has managed to innovate at least a little out of the hole it finds itself in.
Beyond designing and making the chips, Intel has also partnered withDelland Supermicro to work on “co-engineered systems” that are made specifically for the specific AI needs of those companies. For its part, Dell is co-engineering RAG-based solutions leveraging Gaudi 3 and Xeon 6.
Watch out AMD – or not?
AMDhas been the biggest beneficiary of Intel’s largesse in the AI chip space, designing powerful chips that are then manufactured by TSMC. It’s a good business model and one that Intel has largely struggled to replicate or disrupt.
Are you a pro? Subscribe to our newsletter
Sign up to the TechRadar Pro newsletter to get all the top news, opinion, features and guidance your business needs to succeed!
The companyplans to releaseits fifth-gen EPYC Turin 3nm data center chips in the second half of 2024 sometimes, based on its Zen 5 architecture and featuring up to 192 cores and 384 threads, so more than a match for the Xeon and Gaudi chips, at least on paper.
In the closely-watchedTop500 list, the Intel-powered Aurora supercomputer, which is not yet fully operational, came second behind the AMD-powered Frontier, but did take first in an AI benchmark.
MORE FROM TECHRADAR PRO
Max Slater-Robins has been writing about technology for nearly a decade at various outlets, covering the rise of the technology giants, trends in enterprise and SaaS companies, and much more besides. Originally from Suffolk, he currently lives in London and likes a good night out and walks in the countryside.
New fanless cooling technology enhances energy efficiency for AI workloads by achieving a 90% reduction in cooling power consumption
Samsung plans record-breaking 400-layer NAND chip that could be key to breaking 200TB barrier for ultra large capacity AI hyperscaler SSDs
NYT Strands today — hints, answers and spangram for Sunday, November 10 (game #252)
