# MLPerf Inference

> MLPerf Inference: MLCommons' public inference benchmark. Its rules fix the model, dataset and accuracy target; the Server scenario reports the highest…

Canonical: https://llmdojo.dev/glossary/mlperf

Also called MLPerf.

MLCommons' public inference benchmark. Its rules fix the model, dataset and accuracy target; the Server scenario reports the highest Poisson arrival rate that meets the latency limits.

Learn it in the lesson: [Comparing providers and public benchmarks](https://llmdojo.dev/learn/inference-optimization/benchmarks-and-capstone/comparing-providers) (Inference and Model Optimization · Benchmarking Inference, and the Module Capstone).

## Related
- [Warm-up](https://llmdojo.dev/glossary/warm-up)
- [Glossary of every term](https://llmdojo.dev/glossary)