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

Reasoning on ARC Challenge (P@1, P@32, #Tok)

95.5P@1

DeepSeek-V4-pro

19.47639.21358.9578.687Jun 4, 2026Jun 5, 2026Jun 6, 2026Jun 7, 2026Jun 8, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
95.5--
2026.06
93.9--
2026.06
92--
2026.06
91.8--
2026.06
90.4--
2026.06
89.6--
2026.06
88--
2026.06
87.3--
2026.06
87.1--
2026.06
86.8--
2026.06
86.4--
2026.06
85.6--
2026.06
85.5--
2026.06
85.2--
2026.06
84.8--
2026.06
84.7--
2026.06
84.4--
2026.06
84.3--
2026.06
83.7--
2026.06
83.6--
2026.06
83--
2026.06
82.5--
2026.06
82.3--
2026.06
82--
2026.06
81.6--
2026.06
79.3--
2026.06
78.5--
2026.06
76.7--
2026.06
75.0697.44277.8
2026.06
74.7898.12272.4
2026.06
74.0598.21106.2
2026.06
74.0198.89189.4
2026.06
74--
2026.06
73.9--
2026.06
72.8--
2026.06
71.0996.2595.4
2026.06
70.5778.1697.1
2026.06
67.3--
2026.06
65.3--
2026.06
58.8--
2026.06
52--
2026.06
22.7--
2026.06
22.4--