Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Mathematical Reasoning on AIME 25 (# traces, Consistency, Accuracy)

911Trace Count

Uniform

121.64326.57531.5736.43Feb 18, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
9119491.9
2026.02
7639385.6
2026.02
7529488.9
2026.02
6819591.3
2026.02
6189489
2026.02
6109183.3
2026.02
5459686.1
2026.02
4709482.1
2026.02
4109590.2
2026.02
2889789.3
2026.02
2579884
2026.02
2519789.7
2026.02
2239786.7
2026.02
21210083.3
2026.02
21210093.9
2026.02
2119893.3
2026.02
2039790.4
2026.02
19010088.6
2026.02
18110090
2026.02
15210090