Reasoning GYM
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Reasoning-Gym (test)
70.1Accuracy
28
Reasoning-Gym (train)
63.3Accuracy
28
Reasoning Gym (test)
17.8Avg@4
20
Reasoning GYM
8.8Accuracy (Reasoning GYM)
10
Reasoning Gym
43.98Avg@4
8
Reasoning Gym Color Cube
32Accuracy (*)
8
Reasoning Gym Leg Counting
50.67Accuracy
8
Reasoning Gym Zebra Puzzles
39.33Accuracy (*)
8
Reasoning Gym Puzzle24
60.67Success Rate
8
Reasoning Gym Word Sorting
84.33Success Rate
8
Reasoning Gym Spell Backward
97Success Rate
8
Reasoning Gym Manipulate Matrix
8.33Success Rate
8
Reasoning Gym ARC-1D
30.67Success Rate
8
Reasoning Gym
4.2Avg@4 Score
4