Briefcase · AI on Xeon

AI on Xeon starter

This briefcase holds everything you need for a first AI infrastructure conversation: the AMX brief, the Gaudi deck, and the NVIDIA objection card.

AMX inference briefThe brief explains what AMX accelerates, which frameworks it supports, and the customer-ready proof points.PDF · updated Jun 2 · file arrives with R2

Gaudi 3 customer pitch deckThe deck walks through training-scale economics and the cost-per-token story for Gaudi conversations.Deck · updated May 20 · file arrives with R2

"We expect to use NVIDIA for AI"Size the inference load before specifying accelerators.Battlecard · updated Jul 8

Handling the NVIDIA default (role-play)This video records a role-play of the sizing questions that reopen a GPU-only spec.Video · updated Jul 5 · file arrives with R2

Review while this loads

“Your users don't run cores. They run workloads. Every core they quote is a core you license.”