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

Multi-task Knowledge on MMLU-Pro

37.42MMLU-Pro Score

AHD

-1.1638968.85307718.8700528.887023Apr 10, 2026Apr 13, 2026Apr 16, 2026Apr 19, 2026Apr 22, 2026Apr 25, 2026Apr 28, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
37.42133.06
2026.04
36.1123.83
2026.04
35.7134.76
2026.04
35.67187.17
2026.04
35.57256
2026.04
33.9256
2026.04
0.5-
2026.04
0.4867-
2026.04
0.4509-
2026.04
0.4403-
2026.04
0.4389-
2026.04
0.4383-
2026.04
0.4373-
2026.04
0.4261-
2026.04
0.4208-
2026.04
0.3943-
2026.04
0.3811-
2026.04
0.3797-
2026.04
0.3739-
2026.04
0.3701-
2026.04
0.3568-
2026.04
0.3528-
2026.04
0.3454-
2026.04
0.3442-
2026.04
0.3426-
2026.04
0.3422-
2026.04
0.3404-
2026.04
0.3394-
2026.04
0.3383-
2026.04
0.336-
2026.04
0.3354-
2026.04
0.3295-
2026.04
0.3201-