Formal Theorem Proving
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
4.1Proof Length
6
Apr 30, 2026
58.8Classical Score
6
Feb 26, 2026
9.04Pass Rate
6
Feb 26, 2026
35.5Pass@8
5
Jul 2, 2026
17.8Proof Length
5
Apr 30, 2026
100Overall Success Rate
5
Jun 5, 2026
7.3Total Solve Rate
4
Jun 19, 2026
40.6Total Solve Rate
4
Jun 19, 2026
96Accuracy
4
May 25, 2026
92Accuracy
4
May 25, 2026
100Accuracy
4
May 25, 2026
64Accuracy
4
May 25, 2026
56Accuracy
4
May 25, 2026
72Accuracy
4
May 25, 2026
51Accuracy
4
May 25, 2026
26Accuracy
4
May 25, 2026
47Accuracy
4
May 25, 2026
100Accuracy
4
May 25, 2026
1,707.19Average Token Cost
4
May 13, 2026
572.34Average Token Cost
4
May 13, 2026
2,604.24Average Token Cost
4
May 13, 2026
28Proof Length
4
Apr 30, 2026
63Pass@1
3
Feb 26, 2026
29.22Performance Score
3
Feb 26, 2026
19.4Total Solve Rate
2
Jun 19, 2026