Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Multimodal Reasoning on MMBench (English and Chinese Scores)

66MMBench Accuracy (en)

LLaVA-1.5-7B

30.01639.35848.758.042Mar 10, 2025
Updated 3d ago

Evaluation Results

MethodLinks
2025.03
6658.9100
2025.03
64.856.597.9
2025.03
64.357.196.8
2025.03
6153.595.7
2025.03
60.656.7-
2025.03
60.553.995.5
2025.03
58.8--
2025.03
58.248.593
2025.03
57.250.691.8
2025.03
54.747.891.9
2025.03
54.538.1-
2025.03
514890.3
2025.03
48.145.488.8
2025.03
3623.7-
2025.03
31.4--