Symbol-level eval: non-saturated benchmark + model comparison findings#38

Merged

Neverdecel merged 3 commits into

claude/eval-harder-benchmark

Jun 17, 2026

Commits on Jun 17, 2026

feat(eval): symbol-level dataset mining (non-saturated benchmark)
claude
committed
feat(eval): reranker registry, multi-reranker bench, curated symbol set
claude
committed
docs(eval): symbol-level model comparison results and findings
claude
committed