AMD prepares EPYC Venice for kilowatt-class power and cooling

Microloops highlighted custom cold plates and distribution hardware at OCP APAC 2025, revealing AMD plans to cool EPYC ´Venice´ chips with thermal design powers in the 700 to 1400 watt range.

the kilowatt era of CPUs is emerging as AMD prepares specialized cooling for its next-generation SP7 socket powering EPYC ´Venice´. a Microloops presentation at the OCP APAC 2025 Summit focused on system-level cooling with custom high-performance cold plates and cooling distribution units and disclosed that AMD is likely to target CPUs with thermal design powers between 700 and 1400 watts. this development marks AMD´s first step toward a kilowatt-level chip design and the need for corresponding kilowatt-level cooling solutions.

AMD is preparing to roll out the Zen 6 microarchitecture in 2026 with a server-focused flagship called EPYC ´Venice´. the platform is expected to increase core counts significantly, using full-sized Zen 6 cores and an updated I/O subsystem to boost multithreaded throughput. AMD is reportedly configuring Venice to support as many as 256 CPU cores in a single package, and the company projects roughly a 70 percent improvement in multithreaded performance compared to the current EPYC ´Turin´ family.

under the hood, Zen 6 CCDs are planned for TSMC´s 2 nm N2 node, while the server I/O die will introduce next-generation connectivity with PCIe Gen 6 to increase raw device bandwidth for GPUs, NVMe storage, and high-speed NICs. AMD has cited potential memory bandwidth of up to about 1.6 TB/s, which could be achieved through faster DRAM clocks, expanding the memory controller to a 16-channel DDR5 layout from the current 12 channels, or supporting multichannel DIMM formats such as MR-DIMM and MCR-DIMM. taken together, these changes promise much higher multithreaded throughput but also a significant power cost, prompting AMD and ecosystem partners to plan customized cooling solutions for the new platform.

72

Impact Score

Saudi Artificial Intelligence startup launches Arabic LLM

Misraj Artificial Intelligence unveiled Kawn, an Arabic large language model, at AWS re:Invent and launched Workforces, a platform for creating and managing Artificial Intelligence agents for enterprises and public institutions.

Introducing Mistral 3: open artificial intelligence models

Mistral 3 is a family of open, multimodal and multilingual Artificial Intelligence models that includes three Ministral edge models and a sparse Mistral Large 3 trained with 41B active and 675B total parameters, released under the Apache 2.0 license.

NVIDIA and Mistral Artificial Intelligence partner to accelerate new family of open models

NVIDIA and Mistral Artificial Intelligence announced a partnership to optimize the Mistral 3 family of open-source multilingual, multimodal models across NVIDIA supercomputing and edge platforms. The collaboration highlights Mistral Large 3, a mixture-of-experts model designed to improve efficiency and accuracy for enterprise artificial intelligence deployments starting Tuesday, Dec. 2.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.