Event announcement date · Intel
Intel reports MLPerf v6.1 inference gains driven by software optimization
Intel said Xeon 6980P delivered 2.4x higher Llama 3.1 8B server throughput than its v6.0 result on the same processor and socket count, with broader Xeon and Arc Pro participation in MLPerf Inference v6.1.
Why it matters
Original announcement / source · Intel ↗![Banseog [ GLOBAL ]](/assets/brand/banseog-global.svg?v=20260920){kind=small}