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

Multi-domain Knowledge and Reasoning on MMLU-Pro

77.7Accuracy

Qwen3

19.293634.456849.6264.7832Apr 2, 2026Apr 4, 2026Apr 7, 2026Apr 10, 2026Apr 13, 2026Apr 16, 2026Apr 19, 2026
Updated 3mo ago

Evaluation Results

MethodLinks
2026.04
77.72,400
2026.04
77.31,900
2026.04
77.13,400
2026.04
76.83,600
2026.04
76.43,500
2026.04
75.61,500
2026.04
24.17-
2026.04
23.45-
2026.04
22.91-
2026.04
22.18-
2026.04
21.54-