Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Science Question Answering on ARC Easy

96.95Accuracy

Hard-routing MoE

21.435641.040360.64580.2497Nov 21, 2024Feb 5, 2025Apr 23, 2025Jul 9, 2025Sep 23, 2025Dec 9, 2025Feb 24, 2026
Updated 2d ago

Evaluation Results

MethodLinks
2026.02
96.95
2026.02
93.31
2026.02
92.6
2026.02
90.82
2026.02
90.69
2026.02
89.38
2026.02
87.56
2026.02
84.49
2026.02
84.41
2026.02
82.95
2026.02
82.78
2026.01
79.3
2026.01
77.9
2026.01
77.5
2026.02
77.5
2026.01
77.25
2026.01
76.9
2026.02
76.3
2026.01
75.2
2026.01
75.1
2026.01
75.1
2026.01
74.4
2026.02
74.2
2025.12
74.1
2026.02
73.9
2025.12
73.7
2026.02
73.4
2026.01
73.1
2026.01
73.1
2026.01
72.69
2026.01
72.64
2025.12
72.6
2026.01
72.52
2026.02
72.5
2026.01
72.4
2026.02
72.4
2026.01
72.31
2026.01
72.3
2026.01
72.18
2026.01
72.1
2026.01
71.97
2026.01
71.84
2026.02
71.75
2026.02
71.6
2025.12
71.3
2026.01
71.21
2026.01
71.13
2026.02
71
2025.12
70.9
2026.01
70.72
2026.02
69.3
2026.02
69.07
2025.12
68.9
2025.12
68.8
2026.02
68.7
2026.01
68.5
2025.12
68.5
2026.01
66.75
2026.02
65.6
2026.02
63.3
2026.02
63.3
2026.02
62.3
2026.02
61.7
2026.01
61.02
2026.02
58.8
2026.02
57
2026.01
56.61
2026.02
54.4
2024.11
54.08
2024.11
53.54
2024.11
53.45
2024.11
53.03
2024.11
53.03
2026.02
52.79
2024.11
52.69
2024.11
52.65
2024.11
52.36
2024.11
52.02
2024.11
51.56
2024.11
50.67
2026.02
50.6
2026.02
48.53
2024.11
47.18
2024.11
47.18
2026.01
43.73
2026.01
43.56
2026.01
41.16
2026.02
36.9
2026.01
35.1
2026.01
30.81
2026.01
29.5
2026.01
26.81
2026.01
26.6
2026.01
26.52
2026.01
26.35
2026.01
26.14
2026.01
26.05
2026.01
25.72
2026.01
25.63
2026.01
24.34
Showing 100 of 101 rows