swe
Results of cap-evolve runs on SWE-bench: SWE-bench and its variants (Verified, the Red Hat rh-swebench set). One of the per-benchmark result repos in the skillberry-runs org.
On the dashboard: https://skillberry-runs.github.io/dashboard/swe/. The run rows (reward,
cost, models per run) are on the main table (filter
Benchmark); each row's Open UI opens that run's CapEvolve dashboard from this repo's ci_ui/.
What is here
ci_ui/<run_id>__<tier>-<bench>/: per-run "Open UI" data (23 runs, 2026-07-24 to 2026-10-06), moved on 2026-11-17 from cap-evolve'sbenchmark-historybranch. Seeci_ui/README.md.- Write-ups (
*.md) and report pages (ui/*.html) added here are published on the dashboard too.
bench value in the run records |
runs |
|---|---|
swebench |
24 |
swebench-verified |
3 |
rh-swebench |
1 |
Adding to it
See the dashboard's PUBLISHING.md: section B for write-ups, F for "Open UI" data.