PackageTrack
Sign in Get early access

benchmark_harness_plus

A statistically rigorous benchmarking harness with median-based comparisons, CV% reliability metrics, and outlier-resistant measurements.

3.0.1 modulovalue/benchmark_harness_plus

What this package is like to depend on

Last release 7 months ago

09 Jan 2026

Too new to tell

only 1 release windows

Nearly every release is documented

notes for 6 of 6 stable releases

Nothing withdrawn

no release was ever pulled

7 months old

6 releases · first in 2026

6 releases in the last 12 months

see the full history below

Release timeline

6 releases · Jan 2026 to Jan 2026
Release Pre-release

Releases

latest 6
  1. 3.0.1 09 Jan 2026
    Release notes
    • Fix typo in README
    Open source →
  2. 3.0.0 09 Jan 2026
    Release notes
    • Breaking: Added Percentage type for type-safe percentage values
    • Breaking: cv() now returns Percentage instead of double
    • Breaking: BenchmarkResult.cv now returns Percentage
    • Breaking: BenchmarkResult.improvementVs() now returns Percentage
    • Breaking: BenchmarkComparison.improvementPercent renamed to improvement and returns Percentage
    • Breaking: reliabilityFromCV() now takes Percentage instead of double
    Open source →
  3. 2.0.0 09 Jan 2026
    Release notes
    • Bump SDK constraint to ^3.5.0
    • Update extra_pedantic to ^5.0.0
    • Add GitHub Actions CI
    Open source →
  4. 1.2.0 07 Jan 2026
    Release notes
    • Removed GC triggering between variants (can cause more noise than it reduces)
    Open source →
  5. 1.1.0 07 Jan 2026
    Release notes
    • Added "fastest" column to table output showing the minimum (best) sample time
    • Added "Fastest" line to detailed result output
    • Added new example demonstrating when minimum vs median should be used
    • Restructured examples into separate files for easier reference
    Open source →
  6. 1.0.0 06 Jan 2026
    Release notes
    • Initial release
    • Statistical functions: mean, median, stdDev, cv, min, max
    • BenchmarkConfig with quick, standard, and thorough presets
    • BenchmarkResult with computed statistics and reliability assessment
    • BenchmarkComparison for comparing variants
    • Benchmark harness with warmup and randomized ordering
    • Result formatting: table, detailed, and CSV export
    • Comprehensive documentation and examples
    Open source →

Every package, every release, already written down.

The archive is open and free. Watching your own project is what we are building next.

Browse the archive