About github-action-benchmark
github-action-benchmark is a GitHub Action for continuous benchmarking. If your project has benchmark suites, the action reads their output in a workflow, stores the results and helps you watch performance change over time so slowdowns are noticed before release.
It can save results to a branch on GitHub Pages and display them as charts, and it can compare new results with previous ones to detect regressions. When results get worse than a configured threshold, it can raise an alert through a commit comment or fail the workflow. It supports many tools, including cargo bench, go test, benchmark.js, pytest-benchmark, Google Benchmark, Catch2, BenchmarkTools.jl, BenchmarkDotNet, JMH and custom benchmarks, and polyglot repositories can use several at once.
The action is written in TypeScript and licensed under MIT. It runs inside your own GitHub Actions workflows, with example projects and live workflows provided in the repository. It suits maintainers of libraries and performance-sensitive applications who want automated regression tracking in CI.
Key features
- Collects results from many benchmark tools
- Charts results on GitHub Pages
- Regression detection with thresholds
- Alerts via commit comment or workflow failure
- Support for polyglot repositories
- Custom benchmark formats
Good fit for
- Tracking performance of a library over time
- Blocking pull requests that slow down hot paths
- Publishing benchmark history for a project
github-action-benchmark: questions and answers
- What is github-action-benchmark used for?
- github-action-benchmark is a GitHub Action that collects benchmark results, charts them on GitHub Pages and alerts you when performance regresses. It is a good fit for tracking performance of a library over time, blocking pull requests that slow down hot paths and publishing benchmark history for a project.
- Is github-action-benchmark open source?
- Yes. github-action-benchmark is open source under the MIT licence. Its source code is on GitHub at benchmark-action/github-action-benchmark and is written mainly in TypeScript.
- Is github-action-benchmark free?
- Yes. github-action-benchmark is open source, so the software itself is free to use.
- What are some alternatives to github-action-benchmark?
- Similar open-source tools in the Testing & QA category include Bencher, k6 and ClusterFuzzLite. SaaS products in the same category include Codecov, Coveralls and Cypress Cloud.
- Is github-action-benchmark actively maintained?
- Yes. The most recent commit to github-action-benchmark was on 15 September 2026, and the latest release is v1.22.2, published on 15 September 2026. The project has 1.3k stars on GitHub, with 8 added in the last 30 days.
Open-source alternatives to github-action-benchmark
See all
Bencher
Testing & QA
π° Bencher - Continuous Benchmarking
OSSβ 929
k6
Testing & QA
A modern load testing tool, using Go and JavaScript
AGPL-3.0vs OpenText LoadRunnerβ 32k
ClusterFuzzLite
Testing & QA
ClusterFuzzLite - Simple continuous fuzzing that runs in CI.
Apache-2.0β 546
Locust
Testing & QA
Write scalable load tests in plain Python ππ¨
MITvs OpenText LoadRunnerβ 28k
QA Wolf CLI
Testing & QA
QA Wolf from anywhere β your terminal, your CI, your AI agent.
Apache-2.0β 3.5k
TestCafe
Testing & QA
A Node.js tool to automate end-to-end web testing.
MITvs Katalonβ 9.9k
SaaS alternatives to github-action-benchmark
See all
Codecov
Testing & QA
Code coverage reporting and test analytics that comments on pull requests
SaaS
Coveralls
Testing & QA
Hosted code coverage tracking that reports on commits and pull requests
SaaS
Cypress Cloud
Testing & QA
Dashboard and parallelization service for Cypress end-to-end test runs
SaaS
desplega.ai
Testing & QA
No-code AI platform that automates end-to-end QA test creation, running and maintenance
SaaS
Emailens
Testing & QA
Email QA platform that tests rendering across major email clients, with React Email, MJML and Maizzle support
SaaS
loader.io
Testing & QA
Cloud load testing service for web apps and APIs from SendGrid
SaaS

