Benchmark
SimpleBench
Test everyday reasoning, social understanding, and resistance to misleading wording.
Published leaderboard
Explore scores and evaluation settings on the publisher’s leaderboard.
Open leaderboardCoverage
More than 200 text questions covering spatial and temporal reasoning, social intelligence, and trick questions.
Methodology
Answer the benchmark questions and compare accuracy under the stated evaluation settings. The original multiple-choice leaderboard reports average performance across five runs; a public question subset and evaluation code are available.