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

Multimodal Question Answering on MMBench CN

84.6Accuracy

Qwen3-VL-8B

47.36857.03466.776.366Jun 16, 2024Oct 18, 2024Feb 19, 2025Jun 24, 2025Oct 26, 2025Feb 27, 2026Jul 2, 2026
Updated 22d ago

Evaluation Results

MethodLinks
2026.07
84.6
2025.10
81.18
81.01
2025.10
81.01
80.41
2025.08
80.1
2026.07
80
2026.07
79.7
2025.08
79.5
2026.07
79.4
2026.07
79.3
2025.08
79.2
2026.07
78.9
2026.07
78.8
2026.07
78.4
2025.08
77.6
2026.07
77.5
2025.08
77
2026.07
76.6
2025.08
73.5
2025.08
73.3
2025.08
73
2024.06
72.6
2025.08
71.8
2024.06
71.2
2025.08
70.9
2025.08
69.1
2024.06
68.9
2025.08
68.9
2024.06
68.2
2024.06
67.8
2024.06
67.5
2025.08
65.1
2025.08
64.9
2026.03
64.52
2026.03
63.6
2026.03
63.57
2026.03
63.49
2026.03
62.54
2026.03
60.4
2026.03
60.3
2026.03
59.88
2026.03
59.02
2026.03
58.3
2026.06
58.2
2026.03
58.16
2026.06
58.1
2026.03
57.9
2026.03
57.56
2026.06
57.3
2026.06
56.4
2026.06
55.2
2026.06
54.6
2026.06
54
2026.06
53.9
2026.06
53.4
2026.06
52.6
2026.06
52.3
2026.06
50.3
2026.06
49.6
2026.06
48.8