Repository navigation

Public EQ-Bench as its own metric (13→35 models) + Python 3.13 pin - #21

Merged

TechNickAI merged 6 commits into

mainfrom

benchmark/eq-public-discovery

Jul 27, 2026

Commits on Jul 26, 2026
- feat(benchmarks): auto-discover new models + public EQ-Bench as its own metric
  Nick Sullivan
  committed
  bac5329
- fix(benchmarks): remove qwen3.6-plus's inherited EQ score from its predecessor
  Nick Sullivan
  committed
  0239f5c
- style(benchmarks): apply ruff-format to new EQ-Bench module and tests
  Nick Sullivan
  committed
  f3795f1
- feat(benchmarks): add 9 frontier models with real public EQ-Bench scores
  Nick Sullivan
  committed
  120aec3
- Pin model-benchmarks to Python 3.13 via uv venv
  Nick Sullivan
  committed
  4aca944
- ci: match local Python 3.13 pin in GitHub Actions
  Nick Sullivan
  committed
  68bc4e5