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

Multi-image Understanding on MuirBench

74.5Score

Gemini 2.5 Flash

-2.9817.13537.2557.365Dec 6, 2024Mar 1, 2025May 25, 2025Aug 18, 2025Nov 11, 2025Feb 4, 2026Apr 30, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
74.5
2026.04
72
2024.12
68
2026.04
64.4
2024.12
63.5
2024.12
62.7
2024.12
62.3
2026.04
61.9
2024.12
61.1
2026.04
55.8
2024.12
54.8
2024.12
54.4
2024.12
51.2
2024.12
51.1
2024.12
50.6
2024.12
48.7
2024.12
45.2
2025.04
41.8
2026.04
41.8
2026.04
41.3
2025.04
41.1
2026.04
41.1
2025.04
40.9
2026.04
40.9
2025.04
40.7
2026.04
40.7
2024.12
40.6
2024.12
40.5
2026.04
39.2
2025.04
38.6
2026.04
38.6
2024.12
38.5
2026.04
36.9
2025.04
35.1
2026.04
35.1
2024.12
32.5
2024.12
29.9
2024.12
29.3
2024.12
25.5
2025.04
0
2026.04
0