TechWatch

Intel is pursuing a three-track AI strategy

Intel and Nvidia team up in AI data center.

Jon Peddie

Intel is telling a three-chapter story. First, Crescent Island arrives as a memory-rich inference GPU (160GB LPDDR5X) aimed at tokens-at-scale services. Second, Celestial desktop GPUs debut the Xe3 design with more cores and bigger caches, pushing smoother frames and faster creator workflows. Third, a rack-scale twist pairs Gaudi 3 (decode) with Nvidia B200 (prefill) over ConnectX-7, yielding ~1.7× better perf/TCO on small dense models. Wrapped around it: fresh capital (US $8.9B, Nvidia $5B) and a joint CPU-GPU SoC plan to widen Intel’s AI playbook. “I am thrilled to announce a deep collaboration with my good friend Jensen Huang and the
...

Enjoy full access with a TechWatch subscription!

TechWatch is the front line of JPR information gathering service, comprising current stories of interest to the graphics industry spanning the core areas of graphics hardware and software, workstations, gaming, and design.

A subscription to TechWatch includes 4 hours of consulting time to be used over the course of the subscription.

Already a subscriber? Login below

This content is restricted

Subscribe to TechWatch