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

Video Reasoning on TempCompass

86.7Score

GPT-5

25.75641.57857.473.222Oct 19, 2025Nov 30, 2025Jan 12, 2026Feb 23, 2026Apr 7, 2026May 19, 2026Jul 1, 2026
Updated 18d ago

Evaluation Results

MethodLinks
2026.06
86.7
84.3
75
2026.06
74.5
2026.06
73.8
2025.10
73.3
2025.10
73.3
2026.06
73.2
2026.07
73.2
2026.07
73.2
2026.06
72.9
2026.07
72.6
2026.07
72.6
2026.07
72.6
2026.07
72.6
2026.06
72.5
2025.10
72.4
2025.10
72.4
2025.10
72.3
2025.10
72.2
2025.10
72.2
2026.07
72.2
2026.07
72.2
2026.07
72.2
2026.06
71.9
2026.06
71.8
2025.10
71.6
2026.06
69.9
2026.06
69.8
2026.07
66.6
2026.07
66.5
2026.07
66.5
2026.06
64.2
2026.06
62.5
2026.07
62.5
59.7
2026.07
59.7
58.8
2026.07
58.8
2025.10
56.9
2026.07
56.9
2025.10
45.6
2026.07
45.6
2025.10
30.4
2025.10
29.1
2025.10
28.1