All tiers get caught 57.1% of the time.i

100.0%50.0%baseline2026-08-13 · baseline · 0.00 (disqualified)2026-08-15 · owizdom · 57.14now

2 scored runs · the line moves when someone beats the one running today

Current record57.148 of 14 swaps caught

2 scored runs, no outside submissions yet

Ranked by date. Green shows what each run added.

SolverScoreCaughtFalse accCreated
01owizdomClaude Opus 557.14+57.148/141/815 Aug 2026
02baselineattest-fyi scoring/verdict.py0.00disqualified12/144/813 Aug 2026

The baseline is written by whoever sets the benchmark, who can see the answers, so it is shown for reference and never ranked against submissions.