Version 4.1 of LLM’s processing performance measurement software for Artificial Intelligence was released today “MLPerf“. We already saw the first results achieved by the AMD Instinct MI300X together with new generations of AMD processors, and now it is Intel that has used this new version of MLPerf v4.1 to show the performance leap achieved by its new Xeon 6 processors equipped with high-performance P-Cores.
On average the results of a Intel Xeon 6980P “Granite Rapids” with P-Core cores “Redwood Cove” show 87% more performance than a 5th Generation Intel Xeon Platinum 8592+ “Emerald Rapids” based on the previous “Golden Cove” P-Cores architecture.
To do this, the company has carried out various tests with several language models, both on the new Granite Rapids processors and on the previous Emerald Rapids processors.
This means 17 times more performance than the Intel Xeon processors offered four generations ago in AI Inference tasks. Making clear the evolution that Intel has had in this sector as Artificial Intelligence has become popular and deployed on a large scale.
In the following graph you can see how the biggest jump in performance occurred between the third and fourth generation, however, between the fifth and sixth generation of Intel Xeon we can also see a quite an amazing jump which is close to double the performance.
End of Article. Tell us something in the Comments!
Antonio Delgado
Computer Engineer by training, editor and hardware analyst at Geeknetic since 2011. I love to dissect everything that passes through my hands, especially the latest hardware that we receive here to do reviews. In my free time I tinker with 3D printers, drones and other gadgets. For anything you need, here I am.
Add Comment