Recorded votes
—accepted production decisionsTHEME PARK BENCHMARK · LIVE RESULTS
The Theme Park
leaderboard.
Scores come from blind, category-matched production votes. Confidence tiers show when the evidence is still too close to call.
Loading the current ranking snapshot…
CURRENT VIEW
Overall rankings
Models shown
—eligible for this viewEvidence coverage
—model slots at the current thresholdHow to read this: each category is shown independently. A model reaches established status after three decisions in that category; until then, the score remains provisional.
SCORE COMPARISON
See the gap at a glance.
Bars share a 0–100 scale, so the distance between runs is easy to compare. The gap labels measure points behind the current leader.