Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Multi-gate races

Benchmarks

Task NameDataset NameSOTA ResultTrend
Gate passageFull multi-gate races (Held-out)
Gate Passage Success Rate27
4
Gate passageFull multi-gate races (All)
Success Rate99
4
Showing 2 of 2 rows