- AMD EPYC 9006 “Venice” offers mammoth 256-core, 512-thread configuration incorporating up to 1GB L3 cache and PCI-E 6.0 and 16 DDR5 memory channels under the hood
- It’s the first time AMD’s EPYC series sees high-core-count CPUs push past the 5GHz barrier
- AMD claims a win by saying its latest EPYC CPU offers “the most agents per watt, per dollar, per rack”
AMD’s recently announced 6th generation EPYC 9006 “Venice” looks to be a true silicon milestone (the first x86 server CPU on a 2nm-class node, up to 256 cores and 512 threads) that sees the company secure its position in the AI data center CPU market.
The launch comes at a time when AMD has caught Intel on the back foot, but finds its position threatened in turn by Nvidia’s Vera CPU, which it touts as a faster and more efficient alternative to traditional x86 offerings, including AMD’s own EPYC series.
With a staggered launch in place for the four CPUs AMD showcased at the AAI event, the stack is expected to be fully available to consumers sometime in 2027, though its powerful 256-core CPU is targeted for a Q4 2026 launch.
Latest videos fromTechRadar
A massive upgrade on several fronts
AMD’s Venice is a serious generational leap. It’s built on TSMC’s N2 process, the foundry’s first step from FinFET to gate-all-around nanosheet transistors, and Team Red says the family is the first high-performance computing product in the industry to reach volume production at the node.
AMD’s recently announced flagship model, the 256-core, 512-thread EPYC 9996, features its new SP7 socket that offers 16-channel DDR5 memory, PCIe 6.0 support, and boost clocks that push past 5GHz.
This effectively doubles what is currently offered for AMD’s current generation of EPYC CPUs: the memory moves to 16 channels of DDR5-8000 or MRDIMM at 12,800 MT/s for around 1.6 TB/s per socket.
AMD goes one step further with the 9006X series (expected 2H 2027), which replaces the cores it’s cutting back on with a much larger L3 cache per core. core, a design that it puts forward makes it much better than the competition for workflows including HPC, technical computing and data-intensive workloads that benefit from the greater bandwidth cathche.
“6th generation AMD EPYC delivers workload-optimized computing for the full spectrum of modern data center needs; from running multiple agents and feeding GPUs across rack-scale AI infrastructure to powering the business-critical applications and services organizations rely on every day,” noted SVP & GM of Compute and Enterprise AI at AMD, Dan McNamara.
AMD claims that the shape of the data center is changing to a three-tier agent architecture, with each tier creating a distinct CPU demand, resulting in workflows centered around executing the agent sandbox, AI host nodes, and general-purpose servers that all of its EPYC CPUs are designed to handle.
The EPYC 9006 series also powers AMD’s rack-scale Helios as it aims to take on Nvidia’s Vera Rubin NVL72 rack-scale offering with its own solution.
With AMD also offering a more mainstream 9006 SP8 and the EYPC 9006 LP as an AI host node CPU using LPDDR memory, it aims to capture all ends of the AI data center CPU market and cement its current lead in a segment that is sure to become increasingly competitive in the coming days.
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews and opinions in your feeds.
