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

Mathematical Reasoning on AIME-25 (pass@4)

55.3Pass@4

Hybrid-LoRA

-2.21212.71927.6542.581May 12, 2026
Updated 14d ago

Evaluation Results

MethodLinks
2026.05
55.3
2026.05
52.7
2026.05
52
2026.05
48.7
2026.05
47.3
2026.05
45.3
2026.05
44
2026.05
42
2026.05
40.7
2026.05
38.7
2026.05
37.3
2026.05
36.7
2026.05
36.7
2026.05
35.3
2026.05
34
2026.05
33.3
2026.05
32.7
2026.05
32
2026.05
31.3
2026.05
31.3
2026.05
30.7
2026.05
21.3
2026.05
20.6
2026.05
10.7
2026.05
0