Intel is reaffirming its commitment to next generation graphics IP in the data center, countering speculation that the roadmap might stall after Xe3P. A presentation by Anil Nanduri, vice president of GTM and product management for Artificial Intelligence accelerators, confirms that Intel is planning a dedicated graphics portfolio beyond the recently introduced Crescent Island. The new GPU IP, currently labeled Xe Next and still in development without an official name, is positioned to follow the rollout of Xe3P based Crescent Island GPUs designed for inference workloads.
With Xe Next, Intel is preparing a data center strategy built around two distinct production lines. On one side is the inference focused Crescent Island family, derived from Xe3P. On the other is Jaguar Shores, a long anticipated GPU aimed at Artificial Intelligence training and high performance computing workloads. The structure suggests Intel wants to clearly differentiate products optimized for inference from those tuned for large scale training and scientific computation, while keeping both under a unified graphics roadmap.
For Artificial Intelligence training and high performance computing workloads, Intel’s Jaguar Shores design is reportedly being finalized and will be completed by the middle of this year, with production starting shortly after. Assuming a perfect design, we could see Intel’s Jaguar Shores GPU late this year or early 2027. It will likely involve a mix of Intel’s internal production in collaboration with TSMC or Intel’s pure 18A design, assuming yields are manageable. Intel confirmed HBM4 memory will be used in these accelerators, and SK hynix appears to be the main partner supplying Intel with this memory. As a rack scale design, Intel will compete with NVIDIA and AMD in the highest margin business, signaling an aggressive push into the top tier segment of the data center accelerator market.
