Intel unveiled its sixth-generation Xeon server processors on Tuesday, marking a significant push to regain lost ground in the data center market from its rival AMD. This new lineup, alongside the aggressively priced Gaudi 3 artificial intelligence (AI) accelerator chips, is Intel’s latest strategy to reclaim dominance in the competitive chip industry.
The Xeon 6 server processors are critical for Intel, which has seen its market share in the data center sector dwindle. Over the past year, Intel’s share of the x86 data center market has fallen by 5.6 percentage points to 76.4%, while AMD has captured 23.6%, according to Mercury Research. AMD’s gains are partly due to its use of Taiwan Semiconductor Manufacturing Co (TSMC) for chip fabrication, capitalizing on Intel’s manufacturing delays.
Intel’s Xeon 6 processors are available in two main versions: a high-performance model and an “efficiency” model. The efficiency variant is designed to replace older-generation chips and significantly reduce server rack space requirements. Intel claims that achieving the same computing power as its second-generation chips now requires about 67% fewer server racks with the efficiency model. This model is tailored for media serving, website hosting, and database calculations.
The more powerful performance model is geared toward heavy computational tasks, such as generating responses from complex AI models. While the efficiency model is available immediately, the performance model will be released in the third quarter of this year, with additional variations planned for next year.
The launch of the sixth-generation Xeon chips was delayed by a year as Intel opted for a new manufacturing process. This delay has been a point of contention as it allowed competitors like AMD to gain a foothold in the market.
In addition to the Xeon chips, Intel also highlighted its Gaudi 3 AI accelerator chips, which are set to undercut competitors significantly in terms of pricing. An accelerator kit with eight Gaudi 3 chips is priced at approximately $125,000, compared to the earlier Gaudi 2’s list price of $65,000. In contrast, a comparable Nvidia HGX server system with eight H100 AI chips can cost over $300,000, according to custom server vendor Thinkmate. This aggressive pricing positions Gaudi 3 as a more affordable alternative to Nvidia’s high-end AI chips.
Intel first detailed the Gaudi 3 AI chip in April, positioning it as a cost-effective competitor to Nvidia’s offerings. The move is seen as an attempt to disrupt Nvidia’s stronghold in the AI chip market.
On the consumer side, Intel also introduced its next-generation laptop chip, named Lunar Lake, which boasts 40% lower power consumption and a more powerful AI processor. This chip is expected to ship in the third quarter, reflecting Intel’s broader strategy to integrate advanced AI capabilities across its product lines.
As Intel rolls out these new products, the tech giant aims to reassure investors and stakeholders of its capability to innovate and compete effectively in the rapidly evolving semiconductor landscape. The strategic focus on high-efficiency and high-performance chips, along with competitive pricing for AI accelerators, underscores Intel’s commitment to reclaiming its leadership position in the market.