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

Multilingual Language Understanding on MMLU-ProX

78.1Accuracy

Qwen3-8B (Thinking)

27.03640.29353.5566.807Aug 20, 2025Sep 17, 2025Oct 15, 2025Nov 12, 2025Dec 10, 2025Jan 7, 2026Feb 5, 2026
Updated 25d ago

Evaluation Results

MethodLinks
2025.08
78.1
2026.02
72
2025.08
71.1
2026.02
70.1
2025.08
69
2025.08
69
2026.02
66.1
2026.02
65.7
2026.02
64.1
2026.02
60.2
2026.02
57
2026.02
53.3
2026.02
45.3
2026.02
41.8
2026.02
37.9
2026.02
37.7
2026.02
36.5
2026.02
33.4
2026.02
29.5
2026.02
29