Observed arrival · 2026-09-21
Benchable turns benchmark runs into regression verdicts
A self-hostable service for storing arbitrary performance metrics, traces, and benchmark history, with statistically tested regression results.
- For
- Developers and CI maintainers tracking performance regressions
- Worth noticing
- The first metric key creates its own chart and inferred unit; CI can fail on X-Benchable-Regressions.
Field notes
Benchable accepts benchmark runs as structured JSON rather than requiring a fixed schema in advance. A metric such as `build.time_ms` or `api.latency_ms` becomes a tracked series when first encountered, while optional percentile data and spans preserve more context around each run. The example also includes branch, commit, environment, runner, and runtime metadata. A returned regression count and `X-Benchable-Regressions` header are designed to let CI stop a build from the same request.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue