What simply occurred? Intel threw down the gauntlet towards Nvidia within the heated battle for AI {hardware} supremacy. At Computex this week, CEO Pat Gelsinger unveiled pricing for Intel’s next-gen Gaudi 2 and Gaudi 3 AI accelerator chips, and the numbers look disruptive.
Pricing for merchandise like these is often stored hidden from the general public, however Intel has bucked the pattern and offered some official figures. The flagship Gaudi 3 accelerator will value round $15,000 per unit when bought individually, which is 50 % cheaper than Nvidia’s competing H100 knowledge middle GPU.
The Gaudi 2, whereas much less highly effective, additionally undercuts Nvidia’s pricing dramatically. An entire 8-chip Gaudi 2 accelerator equipment will promote for $65,000 to system distributors. Intel claims that is simply one-third the value of comparable setups from Nvidia and different rivals.
For the Gaudi 3, that very same 8-accelerator equipment configuration prices $125,000. Intel insists it is two-thirds cheaper than different options at that high-end efficiency tier.
At (*3*)#Computex2024, Intel CEO @PGelsinger unveiled all new Intel®ï¸Â Xeon®ï¸Â 6 processors, Lunar Lake structure and 80+ new AI PC designs and customary AI kits together with eight Intel® Gaudi® 2 & 3 accelerators. pic.twitter.com/viHlLGQVDd
– Intel India (@IntelIndia) June 7, 2024
To present some context to Gaudi 3 pricing, Nvidia’s newly launched Blackwell B100 GPU prices round $30,000 per unit. Meanwhile, the high-performance Blackwell CPU+GPU combo, the B200, sells for roughly $70,000.
Of course, pricing is only one a part of the equation. Performance and the software program ecosystem are equally essential issues. On that entrance, Intel insists the Gaudi 3 retains tempo with or outperforms Nvidia’s H100 throughout quite a lot of necessary AI coaching and inference workloads.
Benchmarks cited by Intel present the Gaudi 3 delivering up to 40 % quicker coaching instances than the H100 in massive 8,192-chip clusters. Even a smaller 64-chip Gaudi 3 setup gives 15 % larger throughput than the H100 on the favored LLaMA 2 language mannequin, in line with the corporate. For AI inference, Intel claims a 2x pace benefit over the H100 on fashions like LLaMA and Mistral.
However, whereas the Gaudi chips leverage open requirements like Ethernet for simpler deployment, they lack optimizations for Nvidia’s ubiquitous CUDA platform that almost all AI software program depends on at this time. Convincing enterprises to refactor their code for Gaudi could possibly be robust.
To drive adoption, Intel says it has lined up at the least 10 main server distributors – together with new Gaudi 3 companions like Asus, Foxconn, Gigabyte, Inventec, Quanta, and Wistron. Familiar names like Dell, HPE, Lenovo, and Supermicro are additionally on board.
Still, Nvidia is a pressure to be reckoned with within the knowledge middle world. In the ultimate quarter of 2023, they claimed a 73 % share of the info middle processor market, and that quantity has continued to rise, chipping away on the stakes of each Intel and AMD. The client GPU market is not all that totally different, with Nvidia commanding an 88 % share.
It’s an uphill battle for Intel, however these huge worth variations might assist shut the hole.
