M6 Mac mini vs GitHub's hosted macOS runner: an Xcode build benchmark
We wanted a real number for one question. How much faster is a dedicated M6 Mac mini than GitHub's standard hosted macOS runner, on a real iOS app? So we built the open-source Wikipedia iOS app on both, with the same Xcode and the same script, three times each. This post replaces our earlier M4 benchmark post. Every number here comes from runs made on September 24, 2026. Results Median of 3 runs,…
We conducted a benchmark comparing the speed of a dedicated M6 Mac mini against GitHub's standard hosted macOS runner for an open-source Wikipedia iOS app. The test was performed on September 24, 2026, and involved running the same script three times each on the two machines.
The results showed that the M6 Mac mini was approximately twice as fast as the GitHub-hosted runner in compiling the app. Package download times were similar for both machines since they pulled the same packages from GitHub over fast connections. All steps completed successfully on both systems during every run.
When comparing real CI jobs, the dedicated Mac proved to be significantly faster. The median of three runs on the GitHub-hosted runner took 269 seconds, while the M6 Mac mini with warm caches completed the same job in just 27 seconds. This represents a dramatic improvement, with a job that initially took four and a half minutes now finishing under 30 seconds.
The cost difference between the two options is also notable. On GitHub's hosted runner, each job is billed in full minutes, resulting in a cost of about 31 cents per job. However, using the dedicated M6 Mac incurs no additional costs.
The testing was conducted on a project from the Wikimedia repository, specifically the wikimedia/wikipedia-ios app at commit 8691a89, built for the iOS Simulator with the same Swift packages. The incremental build in this test involved adding a single comment line to one Swift file in the app.
While the benchmark focused on package download and build times, it's worth noting that the incremental build took longer than expected due to Xcode's planning process. A more significant change to a widely used file would likely result in even greater time savings on the dedicated Mac.
It's important to consider that the two machines used in the test ran different macOS versions, as each operated on the macOS version provided by its owner at the time of testing. Additionally, the GitHub-hosted runner could potentially cache Swift packages using the actions/cache feature, which would further reduce the download step but not eliminate the clean build step altogether.
In conclusion, the benchmark demonstrates a clear advantage in speed and cost for using a dedicated M6 Mac mini compared to GitHub's standard hosted macOS runner. However, the results should be weighed against the fact that the M6 runner is provided by the company conducting the benchmark.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
