This publication marks the first ever release of results from the MLPerf Storage benchmark, which measures the performance of storage systems in the context of ML training workloads. MLPerf Inference v3.1 Introduces New LLM and Recommendation BenchmarksMLPerf Inference v3.1 includes record participation, with over 13,500 performance results and up to 40% performance gains, from 26 different submitters, and over 2000 power results. In particular, MLCommons® would like to congratulate first time MLPerf Inference submitters Connect Tech, Nutanix, Oracle, and TTA. The MLCommons MLPerf Inference benchmark suite measures how fast systems can run models in a variety of deployment scenarios. First Results for New MLPerf Storage BenchmarkThe MLPerf Storage Benchmark Suite is the first open-source AI/ML benchmark suite that measures the performance of storage for ML training workloads.

Source: CNN September 11, 2023 16:02 UTC