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

Mathematical Reasoning on Beyond AIME

22Pass@1

e3-1.7B + InT+ RL

-0.71365.183211.0816.9768Dec 30, 2025Jan 6, 2026Jan 13, 2026Jan 21, 2026Jan 28, 2026Feb 4, 2026Feb 12, 2026
Updated 3mo ago

Evaluation Results

MethodLinks
2026.01
22
2026.01
21.75
2026.01
20.88
2026.02
19.7
2026.02
19.5
2026.02
17
2026.02
14.6
2026.02
13.9
2026.02
13.7
2026.02
13.1
2026.02
12.6
2026.02
9
2026.02
7.5
2025.12
0.54
2025.12
0.52
2025.12
0.47
2025.12
0.46
2025.12
0.43
2025.12
0.4
2025.12
0.3
2025.12
0.16