Intel Xeon 6980P vs. AMD EPYC 9755 128-Core Showdown With The Latest Linux Software For EOY2025

Written by Michael Larabel in Processors on 17 December 2025 at 11:00 AM EST. Page 3 of 9. 11 Comments.

With the OpenVINO AI toolkit developed by Intel as open-source, AMD EPYC 9005 series perform really great here ever since the AVX-512 introduction in the EPYC 9004 series.

👁 OpenVINO benchmark with settings of Model: Weld Porosity Detection FP16-INT8, Device: CPU. Xeon 6980P 2P was the fastest.

👁 OpenVINO benchmark with settings of Model: Machine Translation EN To DE FP16, Device: CPU. Xeon 6980P 2P was the fastest.

👁 OpenVINO benchmark with settings of Model: Handwritten English Recognition FP16-INT8, Device: CPU. Xeon 6980P 2P was the fastest.

The Intel Xeon 6980P is able to demonstrate one of its few advantages over Turin here though: Advanced Matrix Extensions (AMX). With OpenVINO able to leverage Intel's AMX instructions, for some of these benchmarks there were very nice gains here over the EPYC 9755 performance. Rather staggering in some cases thanks to AMX. Plus the memory bandwidth of MRDIMM-8800 modules helping out too but at higher power use.

👁 OpenVINO benchmark with settings of Model: Vehicle Detection FP16-INT8, Device: CPU. EPYC 9755 2P was the fastest.

👁 OpenVINO benchmark with settings of Model: Road Segmentation ADAS FP16-INT8, Device: CPU. EPYC 9755 2P was the fastest.

👁 OpenVINO benchmark with settings of Model: Person Re-Identification Retail FP16, Device: CPU. Xeon 6980P 2P was the fastest.

👁 OpenVINO benchmark with settings of Model: Person Re-Identification Retail FP16, Device: CPU. EPYC 9755 2P was the fastest.

👁 OpenVINO benchmark with settings of Model: Noise Suppression Poconet-Like FP16, Device: CPU. Xeon 6980P 2P was the fastest.

Overall though there was healthy competition between Turin and Granite Rapids with OpenVINO. Even with AMX and MRDIMMs, the Xeon 6980P processors were not the clear-cut winner here on this up-to-date Linux software stack.

👁 oneDNN benchmark with settings of Harness: Deconvolution Batch shapes_1d, Engine: CPU. EPYC 9755 2P was the fastest.

👁 oneDNN benchmark with settings of Harness: IP Shapes 3D, Engine: CPU. Xeon 6980P 2P was the fastest.

👁 oneDNN benchmark with settings of Harness: Recurrent Neural Network Training, Engine: CPU. EPYC 9755 2P was the fastest.

👁 oneDNN benchmark with settings of Harness: Recurrent Neural Network Inference, Engine: CPU. EPYC 9755 2P was the fastest.

With Intel's oneDNN library there were some Xeon 6980P wins thanks again to AMX but also a number of the benchmarks where the AMD EPYC 9755 performance was decisively better.

11 Comments - Next Page

Page: 1 2 3 4 5 6 7 8 9 Next Page

URL: https://www.phoronix.com/review/xeon-6980p-epyc-9755-2025/3

⇱ Intel Xeon 6980P vs. AMD EPYC 9755 128-Core Showdown With The Latest Linux Software For EOY2025 Review - Phoronix

Intel Xeon 6980P vs. AMD EPYC 9755 128-Core Showdown With The Latest Linux Software For EOY2025