# Statekit Benchmarks
This directory contains benchmark results and documentation for Statekit.
Statekit uses Criterion to benchmark machine construction and common query operations across different machine sizes.
## Current Baseline
The current benchmark baseline is:
- [Statekit v0.3 Benchmark Baseline](baseline-v0.3.md)
The baseline contains the benchmark environment, methodology, measurements, interpretation, and known limitations for Statekit v0.3.
## Benchmark Coverage
The benchmark suite measures:
- transition iteration
- transition membership queries
- state membership queries
- outgoing-target queries
- source projection
- state projection
- machine construction and destruction
Measurements are collected using machines containing 100, 1,000, 10,000, and 100,000 transitions.
## Benchmark Source
The benchmark implementation used for the baseline is available here:
- [Statekit benchmark source](../../benches/machine_queries.rs)
When viewing a release tag, the benchmark source and baseline documentation correspond to that version of Statekit.
## Running the Benchmarks
Run the benchmark suite with:
cargo bench
Benchmark timings depend on hardware and execution conditions. The recorded baseline is primarily intended as a reference point for comparing Statekit performance across implementations.