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

Mathematical Problem Solving on MATH 500

92MATH-500 Accuracy

S2: S1 → SFT(2) 60B Targeted (Reg|Lklhd–26)

-0.8223.277547.37571.4725Apr 21, 2026Apr 25, 2026Apr 30, 2026May 5, 2026May 10, 2026May 15, 2026May 20, 2026
Updated 13d ago

Evaluation Results

MethodLinks
2026.04
9288.382
2026.04
90.889.682.5
2026.04
89.88983
2026.05
88.1--
2026.05
73.1--
2026.05
72.8--
2026.05
64.6--
2026.05
62.1--
2026.05
51.6--
2026.05
6.33--
2026.05
6.25--
2026.05
6.17--
2026.05
5.83--
2026.05
5.58--
2026.05
5.08--
2026.05
5.08--
2026.05
4.67--
2026.05
4.42--
2026.05
2.75--