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

Multimodal Large Language Model Evaluation on MLLM Evaluation Suite

56.7Average Score (All)

SigLIP 2 (giant) + PIVOT

34.44440.2224651.778Oct 18, 2025
Updated 5d ago

Evaluation Results

MethodLinks
2025.10
56.768.554.754.249.3
2025.10
55.668.153.952.448.1
2025.10
55.467.452.853.148.5
2025.10
54.666.952.251.747.7
2025.10
53.966.550.851.946.4
53.667.348.552.546
2025.10
53.267.746.851.746.6
2025.10
52.466.246.650.646.1
2025.10
52.266.545.250.846.3
2025.10
51.465.944.649.145.9
2025.10
50.965.442.349.846
2025.10
49.564.637.848.647.1
2025.10
49.464.541.546.545.1
2025.10
46.362.135.14345
2025.10
43.662.118.749.244.3
2025.10
40.958.417.645.142.6
2025.10
39.752.518.243.344.6
2025.10
37.747.318.140.345.1
2025.10
37.547.417.64144.1
2025.10
36.847.617.340.242
2025.10
35.544.617.238.242.1
2025.10
35.342.517.139.642.1