Intel's next-gen AI architecture is reportedly under preparation by the company, as the respective 'testing vehicle' surfaces online, revealing a rather elegant design.
Intel's Jaguar Shores Would Be The Firm's First Rack-Scale Solution, Using 18A & HBM4
Team Blue's business in the AI segment hasn't been impressive, given that the firm hasn't managed to offer competitive solutions in the market. The last chip we saw from the company was the Gaudi 3 AI accelerator, which barely saw adoption relative to the likes of NVIDIA and AMD. Apart from this, Intel, under CEO Lip-Bu Tan, also cancelled its highly anticipated Falcon Shores AI GPU lineup, instead announcing that the firm will rely on Jaguar Shores. And now, the latter lineup has smiled for the camera for the first time.
Andreas Schilling shared an image of the Jaguar Shores testing vehicle at X, which, according to him, is currently being used by the thermal engineering team, likely to figure out adequate cooling solutions. It is apparently placed on a development board, and the chip itself comes with a massive package size, reportedly measuring 92.5mm x 92.5mm, which indicates that it is an HPC-oriented platform. Moreover, the test vehicle shows the presence of four different tiles and eight HBM sites.
For now, details around Intel's next-gen AI chip architecture are uncertain, but the firm did reveal that its next product would go up to a rack-scale solution, which means Jaguar Shores could be Team Blue's first AI platform that could rival competing options. Moreover, we recently reported on Intel's plans to use SK Hynix's HBM4 memory on Jaguar Shores, which will be a multi-domain and multi-IP product. Jaguar Shores racks will be combined with Intel's upcoming Diamond Rapids Xeon CPUs.
Intel's plans for the AI segment are unclear at this time, considering that the firm is navigating its way around financial and political difficulties. The priority right now is to reduce operating losses and focus on core businesses, but we'll have to wait and see whether AI computational sees interest from the company.
About the author: Muhammad Zuhair is a hardware and technology reporter for Wccftech, specializing in the semiconductor industry and the complex interplay between technology, manufacturing, and geopolitics. His coverage focuses on the corporate strategies and technological roadmaps of industry giants like TSMC, NVIDIA, Samsung, and Intel. Zuhair's expertise lies in deconstructing complex topics such as fabrication nodes (e.g., 2nm process), the economic impact of policies like the CHIPS Act, and the strategic development of AI infrastructure from NVIDIA, AMD and Intel.
Follow Wccftech on Google to get more of our news coverage in your feeds.
