Rankings and lists, about 7 Apr 2027
MLPerf Inference results
MLCommons' spring round of AI inference benchmarks, comparing how fast chips and clouds run standard models.
MLCommons publishes MLPerf Inference results twice a year, in spring and late summer. Each round fixes a set of models, such as large language models, image generators and recommendation systems, along with accuracy targets and test scenarios, so that submitted systems can be compared. Chipmakers, server makers and clouds take part; results show both hardware and how well each submitter tuned its software. The spring round usually appears in early April, so the date here is an estimate.
Providers tied to MLPerf Inference results
Still to be written: Google Cloud Platform.