Automated Theorem Proving
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
99.6Success Rate
100
Jun 18, 2026
34Accuracy
18
May 26, 2026
31.97Accuracy
18
May 26, 2026
0.12Output Tokens (Thousands)
12
Apr 30, 2026
30Success Rate
9
Feb 26, 2026
0.4Output Tokens (k)
6
Apr 30, 2026
0.66Output Tokens (Thousands)
6
Apr 30, 2026
0.16Thousands of Output Tokens
6
Apr 30, 2026
0.08Output Tokens (Thousands)
6
Apr 30, 2026
77.6Proof Success Rate
6
Mar 23, 2026
69.8Proof Success Rate
6
Mar 23, 2026
89Proof Success Rate
6
Mar 23, 2026
79.8Proof Success Rate
6
Mar 23, 2026
56.5Performance
6
Feb 26, 2026
4Solved Problems (Pass@32)
5
Apr 20, 2026
215Solved Problems (Pass@32)
5
Apr 20, 2026
6,235Proof Lines Count
5
Mar 23, 2026
33Pass Rate
5
Mar 12, 2026
98Pass Rate
5
Mar 12, 2026
10Solved Problems (Pass@32)
4
Apr 20, 2026
64Solved Proofs
4
Feb 26, 2026
88Solved Proofs
4
Feb 26, 2026
10Total Solved (Pass@32)
3
Apr 20, 2026
43Solved Problems (Pass@32)
3
Apr 20, 2026
5Solved Problems Count
3
Jun 5, 2026