Legal Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
0.482GPT-5 Grader Score
4
Jun 25, 2026
31.5GPT-5 Grader Score
4
Jun 25, 2026
44.1GPT-5 Grader Score
4
Jun 25, 2026
70.1Top-1 Accuracy
4
May 27, 2026
0.919Hypervolume
4
Mar 12, 2026
3.1Correctness
2
Jun 30, 2026
82.3Accuracy
2
Jun 4, 2026