Formal Theorem Proving
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
100Pass@1
157
Jun 19, 2026
100Success Rate
68
Jun 17, 2026
668Solved Count
56
Jun 12, 2026
57Success Rate
41
Feb 26, 2026
51.2Success Rate
39
Feb 26, 2026
24.26Accuracy
26
May 26, 2026
75Pass@8
24
May 22, 2026
81.1Pass@8
24
May 22, 2026
48Solve Rate
15
Apr 30, 2026
42.2Pass@1
15
Feb 26, 2026
73Pass@1
14
Feb 26, 2026
2.51PutnamBench
13
Apr 30, 2026
3.1567NEQ
13
Apr 30, 2026
66.31Proof Success Rate
12
May 13, 2026
228.64Average Token Cost
12
May 13, 2026
110Proof Lines
12
Mar 24, 2026
44.62Pass@1
12
May 22, 2026
462Solved Problems Count
11
Mar 18, 2026
66.7Ineq-Comp (Seed)
9
May 22, 2026
62.6Pass@1
9
Feb 26, 2026
34.44Accuracy
8
Apr 23, 2026
80Solve Rate
7
Mar 12, 2026
11.2Proof Length
6
Apr 30, 2026
14.5Proof Length
6
Apr 30, 2026
10Proof Length
6
Apr 30, 2026