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

Multimodal Evaluation on Combined Benchmark Suite (GQA, MMB, MME, VQA-T, SQA-I, VQA-v2)

100Relative Accuracy

LLaVA-NeXT-7B

87.72890.91494.197.286Mar 18, 2025
Updated 5d ago

Evaluation Results

MethodLinks
2025.03
100
2025.03
100
2025.03
99
2025.03
99
2025.03
99
2025.03
99
2025.03
97.6
2025.03
97.6
2025.03
97.4
2025.03
97.4
2025.03
97.1
2025.03
97.1
2025.03
97.1
2025.03
97.1
2025.03
96.2
2025.03
96.2
2025.03
95.8
2025.03
95.8
2025.03
95.4
2025.03
95.4
2025.03
93.8
2025.03
93.8
2025.03
92.6
2025.03
92.6
2025.03
92.3
2025.03
92.3
2025.03
88.2
2025.03
88.2