This is the repository containing results and code for the v2.1 version of the MLPerf™ Inference benchmark.
For benchmark code and rules please see the GitHub repository.
Additionally, each organization has written approximately 300 words to help explain their submissions in the MLPerf™ Inference v2.1 Results Discussion.