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

Multimodal Instruction Following on MIA-Bench

8.86Score

GPT-4o

4.99125.995678.0044Oct 8, 2024
Updated 26d ago

Evaluation Results

MethodLinks
2024.10
8.86-
2024.10
8.76-
2024.10
8.43-
2024.10
8.07-
7.94-
2024.10
7.7-
2024.10
7.63-
7.6-
2024.10
7.56-
2024.10
7.54-
7.06-
2024.10
5.14-
2025.10
-66.33
2025.10
-25.25
2025.10
-29.66
2025.10
-62.33
2025.10
-29.79
2025.10
-36.19
2025.10
-35.03
2025.10
-34.85
2025.10
-42.9
2025.10
-34.07
2025.10
-38.54
2025.10
-60.02