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

EXPLORE-Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Long-horizon reasoningEXPLORE-Bench Full
Sobj Score65.59
6
Long-horizon reasoningEXPLORE-Bench Long
Sobj Score59.37
6
Long-horizon reasoningEXPLORE-Bench Medium
Sobj Score66.71
6
Long-horizon reasoningEXPLORE-Bench Short
Object Score66.12
6
Showing 4 of 4 rows