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

Reasoning on ARC Easy

96.63Accuracy

GPT-4

67.218874.854482.4990.1256May 28, 2020May 15, 2021May 2, 2022Apr 19, 2023Apr 5, 2024Mar 23, 2025Mar 11, 2026
Updated 19d ago

Evaluation Results

MethodLinks
2023.11
96.63---
2023.11
93.73---
2026.01
93.43---
2023.11
92.85---
2020.05
92---
2026.01
92---
2026.01
91---
2026.01
90.1---
2026.01
89---
2023.11
87.79---
2026.01
87.54---
2026.01
86.5---
2023.11
86.24---
2023.11
85.31---
2023.11
85.1---
2026.01
85---
2026.01
85---
2026.02
85---
2026.01
84.6---
2026.01
84---
2026.01
84---
2026.01
83---
2026.01
83---
2026.01
83---
2025.05
82.28---
2023.11
82.2---
2025.05
82.11---
2025.05
81.75---
2025.05
81.75---
2026.03
81.1---
2025.03
80.9-79.6-
2026.03
80.9---
2026.03
80.8---
2026.03
80.8---
2023.11
80.68---
2025.05
80.53---
2025.03
80.1---
2023.02
80---
2026.01
80---
2025.03
79.9---
2025.03
79.7---
2026.01
79---
2026.01
79---
2023.02
78.9---
2025.03
78.8---
2025.05
78.77---
2026.01
77---
2026.01
76.8---
2023.02
76.6---
2023.09
76.3---
2023.11
76.26---
2026.01
76.14---
2023.09
76.1---
2025.03
76-74.5-
2026.02
76---
2026.01
75.96---
2026.01
75.8---
2023.09
75.6---
2026.01
75.42---
2023.09
75.4---
2023.02
75.2---
2026.02
75---
2023.09
74.9---
2023.02
74.8---
2025.03
74---
2026.01
74---
2026.02
74---
2025.03
73.7---
2026.01
73.32---
2025.03
72.9---
2026.01
72.85---
2023.02
72.8---
2026.01
72.31---
2026.01
72.14---
2026.02
72---
2023.09
71.9---
2025.05
71.75---
2026.01
71.68---
2025.05
71.58---
2026.01
71.38---
2020.05
71.2---
2026.01
71.13---
2026.01
71.13---
2026.01
70.83---
2025.03
70.5---
2020.05
70.1---
2025.03
70.1---
2026.02
70---
2022.05
69.8---
2026.01
69.65---
2026.01
69.4---
2025.06
69.23---
2026.01
69.1---
2026.01
69---
2023.11
68.98---
2020.05
68.8---
2023.02
68.8---
2025.06
68.43---
2025.06
68.35---
2025.06
68.35---
Showing 100 of 215 rows