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

Visual Instruction Following on LLaVA-Bench Wild

102.3Score

AutoV

30.74849.32467.986.476Dec 8, 2023Mar 10, 2024Jun 11, 2024Sep 12, 2024Dec 14, 2024Mar 17, 2025Jun 19, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.06
102.3-
2025.06
100.6-
2025.06
98.2-
2025.06
98-
2025.06
97.5-
2025.06
85.3-
2025.06
81.9-
2024.01
81.8-
2025.06
80.8-
2025.06
80.6-
2025.06
79.5-
2023.12
79.1-
2023.12
78.6-
2023.12
76.9-
2024.09
76.7-
2024.09
76.6-
2024.09
76.2-
2025.06
76.2-
2025.06
75.7-
2025.06
75.4-
2025.06
74.6-
2024.03
74.4-
2024.09
74.2-
2024.01
73.9-
2024.01
72.6-
2024.03
71.6-
2024.03
71.2-
2025.06
70.8-
2024.01
70.7-
2024.09
69.7-
2024.09
69.6-
2025.06
69.5-
2023.12
68.6-
2024.09
68.6-
2025.06
68.6-
2023.12
68.2-
2024.01
67.7-
2023.12
67.7-
2025.06
67.6-
2024.03
67.3-
2025.06
67.2-
2025.06
66.9-
2025.06
66.6-
2025.06
65.4-
2024.03
63.8-
2024.01
63.7-
2025.06
63.7-
2024.01
63-
2025.06
62.2-
2024.01
60.9-
2024.03
58.2-
2024.01
56.9-
2025.06
55.6-
2024.01
53.8-
2023.12
43.3-
2023.12
40.6-
2024.01
38.1-
2024.03
38.1-
2023.12
38.1-
2024.01
33.5-
2024.08
-69.1
2024.08
-79.9
2024.08
-73.3
2024.08
-38.1
2024.08
-80.1
2024.08
-60.9
2024.08
-58.2
2024.08
-63.4
2024.08
-70.7
2024.08
-81.6
2024.08
-87.3
2024.08
-79.2
2024.08
-86.8
2024.08
-76.4