Interactive Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
63.91Success Rate
31
Feb 26, 2026
47.9Average Reasoning Length (tokens)
18
Mar 13, 2026
34.9Zork1 Score
7
Feb 26, 2026
0.586Success Rate
7
Feb 26, 2026
84.7Score
6
Mar 19, 2026
85Success Rate
2
Feb 26, 2026
54Score
1
Mar 19, 2026
85Human Success Rate
1
Feb 26, 2026