Intel and pals cram 36,864 CPU cores into a 100kW rack while chasing the agentic AI dragon
Jump to main content
Search
REG AD
AI + ML
Intel and pals cram 36,864 CPU cores into a 100kW rack while chasing the agentic AI dragon
Meanwhile, Intel and SambaNova's disaggregated inference blueprint lands its first customer
Tobias Mann
Tobias<br>Mann
Systems editor
Published<br>tue 2 Jun 2026 // 10:37 UTC
COMPUTEX 2026 Intel is working with Foxconn and other infrastructure providers to develop rack-scale reference designs based on the chipmaker’s Xeon processors.<br>Announced during Intel’s Computex keynote on Tuesday, these blueprints aim to provide greater CPU compute densities for running AI agents at scale.<br>While AI models predominantly run on GPUs and other AI accelerators, the agent harnesses, like OpenClaw, which are used to connect them to tools, terminal shells, code interpreters, and other APIs, still run on CPUs.
REG AD
“Our customers are asking us to think at the system level to help them serve real agentic workloads at scale,” Intel CEO Lip Bu Tan said.
REG AD
On stage, Tan revealed two examples of these blueprints. One is aimed at latency-sensitive agentic workloads and another designed for maximum density.<br>Both designs support up to 128 of either Intel’s 128-core Granite Rapids Xeon 6 or 288-core Clearwater Forest Xeon 6+ processors, totaling between 16,384 P-cores and 36,864 E-cores, alongside up to 384 TB of DDR5 in a 100kW power envelope.<br>The reference designs come just months after Nvidia announced a similar rack-scale CPU platform packing 256 of its 88-core Vera CPUs.<br>Arm is also working on a pair of rack-scale reference designs for agentic workloads based on its new AGI CPUs: a 36 kW air-cooled system with 8,160 cores and a 200 kW liquid cooled rack with 45,696 cores.<br>Tan expects systems based on these reference designs to be broadly available from its ODM and OEM partners.
MORE CONTEXT
If cores are what agents crave, Intel's new Clearwater Xeon 6+ might just quench their thirst
Broadcom's custom ASIC biz adds South Korea's FuriosaAI to its empire
GPU behemoth Nvidia on track to be world's leading CPU supplier too, says CFO
AMD puts out new slottable GPU for AI-curious enterprises
Alongside agentic AI workloads, the company also revealed that newly launched inference cloud provider Vector Core Compute will be among the first to deploy the platform, and that Together.AI is its first commercial customer.
The approach is based on Intel's earlier disaggregated AI blueprint it co-developed with partner SambaNova. The architecture desegregates compute heavy prefill operations to Nvidia GPUs while using SambaNova’s AI accelerators for bandwidth-intensive decode operations to boost per-user token output by between 2-3x.<br>If that sounds familiar it’s not dissimilar to what Nvidia is doing with Groq’s LPUs or what AWS is doing with Trainium and Cerebra's waferscale AI accelerators.®
computex 2026<br>cpu<br>intel<br>datacenter<br>ai and ml<br>ai
REG AD
SPONSORED LINKS<br>Building the New Trust Architecture for AI - June 4, 10am PT
CxO
Remote work – not AI – is killing job prospects for the youth
Young professionals may be perfectly productive while working from home, says the New York Fed, but the quality of their output isn't so great, so companies don't want to hire them
Networks
Marvell enters the AI network fray with 102.4 Tbps switch silicon
High radix, low latency and low power is what AI datacenters crave, the chipmaker says
PARTNER CONTENT
AI and data sovereignty in Postgres: An answer to the datacenter energy crisis
A billion AI agents walk into a power grid
Enhanced performance for server consolidation with Intel Xeon 6+
SPONSORED POST: How Intel’s first 18A data center CPU delivers efficiency and TCO gains, with Intel's Kira Boyko
SaaS
AWS reportedly to tuck Elon Musk's Grok into Bedrock, despite zero enterprise demand
The energy drink of frontier models
Security
Russian spy agency says foreign spies turned officials' smartphones into surveillance devices
FSB claims large-scale snoop op compromised phones of senior officials, but gives no technical evidence to back allegations
MOST POPULAR
AI + ML
Netflix wiz creates app to slash AI bills, then open sources it
Security
Disgruntled 0-day hunter 'humiliated' by Microsoft pledges 'bone shattering drop' as Redmond calls cops
Security
Troops’ phones gave away location data to foreign adversaries
AI + ML
Google has seriously leaned into AI enshittification lately
Security
Anthropic to release Mythos-class models to the public
EVENTS
Overcoming the trade-offs in data sovereignty
What does data sovereignty actually mean for your network, which trade-offs are unavoidable? Learn more.
From Prompt to Exploit: How LLMs Are Changing API Attacks
Modern applications are API-driven, interconnected, and often over-permissioned, making them an ideal target for AI-assisted attacks.
Architecting the Future: Unlocking Enterprise Data...