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

Mathematical Reasoning Accuracy on OlympiadBench (test)

57.9Accuracy

MiMo-VL-7B-SFT-2508

0.38815.31930.2545.181Feb 17, 2025May 5, 2025Jul 22, 2025Oct 8, 2025Dec 24, 2025Mar 12, 2026May 29, 2026
Updated 2d ago

Evaluation Results

MethodLinks
57.9-
57.9-
2026.05
39.2-
2026.05
39.2-
2026.05
37.5-
2026.05
35.4-
2026.05
34.4-
2026.05
32.5-
2026.05
32.1-
2026.05
30.4-
2026.05
30.3-
2026.05
29-
2026.05
28.7-
2026.05
28-
2026.05
27.9-
2026.05
26.3-
2026.05
24.5-
2026.05
23.7-
2026.05
20.9-
2026.05
20.5-
2026.05
19.9-
2026.05
19.2-
2026.05
18.8-
2026.05
16.1-
2026.05
16.1-
2026.05
15.8-
2026.05
13.4-
2026.05
11.9-
2026.05
9.6-
2026.05
8.6-
2026.05
6.9-
2026.05
5.086-
2026.05
5.038-
2025.02
4.5-
2026.05
4.328-
2026.05
4.2-
2025.02
4.1-
2025.02
3.8-
2025.02
3.5-
2025.02
2.6-
2026.05
-38.58
2026.05
-43.3
2026.05
-43.25
2026.05
-44.49
2026.05
-43.49
2026.05
-42.47
2026.05
-43.14
2026.05
-34.34
2026.05
-44.35
2026.05
-42.28
2026.05
-46.8
2026.05
-47
2026.05
-34.8
2026.05
-34.4
2026.05
-34.7
2026.05
-32.4
2026.05
-34.9
2026.05
-35.7
2026.05
-34.6
2026.05
-42
2026.05
-42.6
2026.05
-38.6
2026.05
-41.7
2026.05
-43