Repository navigation

feat: explain benchmark model choices - #441

Merged

mattrossman merged 19 commits into

mainfrom

mattrossman/ai-1285-explain-why-each-benchmark-experiment-uses-its-model-and

Nov 11, 2026

Merged

feat: explain benchmark model choices#441
mattrossman merged 19 commits into
mainfrom
mattrossman/ai-1285-explain-why-each-benchmark-experiment-uses-its-model-and

Commits on Oct 26, 2026
- chore(evals): align Claude Code benchmark presets with current defaults
  mattrossman
  committed
  92cea6b
- chore: refresh eval results
  github-actions[bot]
  committed
  24491af
- Revert "chore: refresh eval results"
  mattrossman
  committed
  6d1d412
- chore(evals): bump pinned Claude Code CLI to 2.1.310
  mattrossman
  committed
  47cb1b7
- chore: refresh eval results
  github-actions[bot]
  committed
  1d09dd8
- Revert "chore: refresh eval results"
  mattrossman
  committed
  fa8111e
- style(evals): format opencode runner test
  mattrossman
  committed
  0ec0de6
- feat(evals): explain each benchmark experiment's model and effort on the results page
  mattrossman
  committed
  3ec9967
Commits on Oct 28, 2026
- refactor: keep benchmark rationales in website catalog
  mattrossman
  committed
  e27d80a
Commits on Nov 9, 2026
- chore: sync benchmark rationales with main
  mattrossman
  committed
  01632fc
- refactor: link benchmark rationale claims inline
  mattrossman
  committed
  f7c2341
- style: clarify benchmark rationale hierarchy
  mattrossman
  committed
  e3cdcc8
- style: align source review dates
  mattrossman
  committed
  3b210aa
- fix: scope prerender count to results rows
  mattrossman
  committed
  bde2808
- refactor: attach rationales to published experiments
  mattrossman
  committed
  67d9dc7
- feat: show experiment upload dates
  mattrossman
  committed
  2809237
- style: trim experiment section introduction
  mattrossman
  committed
  bf27a76
- fix: cite Codex model recommendation
  mattrossman
  committed
  410bc4e
Commits on Nov 11, 2026
- fix: keep experiment upload dates visible
  mattrossman
  committed
  2d6b884

Commits

Commits on Oct 26, 2026

Commits on Oct 28, 2026

Commits on Nov 9, 2026

Commits on Nov 11, 2026