Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multimodal Question Answering on MMBench-CN (dev)
Loading...
73
Accuracy
Qwen
60.208
63.529
66.85
70.171
Aug 1, 2025
Accuracy
Average Relative Performance
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Average Relative Performance
Qwen
Backbone=Qwen2.5-VL-3B...
2025.08
73
100
HiPrune
Backbone=Qwen2.5-VL-3B...
2025.08
71.4
98.7
HiPrune++
Backbone=Qwen2.5-VL-3B...
2025.08
71.1
98.7
FastV
Backbone=Qwen2.5-VL-3B...
2025.08
70.6
97.4
VisionZip
Backbone=Qwen2.5-VL-3B...
2025.08
69.8
97.7
HiPrune++
Backbone=Qwen2.5-VL-3B...
2025.08
69.4
97.1
FastV
Backbone=Qwen2.5-VL-3B...
2025.08
69.2
95.9
HiPrune
Backbone=Qwen2.5-VL-3B...
2025.08
69.1
97.1
VisionZip
Backbone=Qwen2.5-VL-3B...
2025.08
67.4
96.2
HiPrune
Backbone=Qwen2.5-VL-3B...
2025.08
65.4
93
HiPrune++
Backbone=Qwen2.5-VL-3B...
2025.08
64.7
93
VisionZip
Backbone=Qwen2.5-VL-3B...
2025.08
63.3
91.5
FastV
Backbone=Qwen2.5-VL-3B...
2025.08
60.7
86.4
Feedback
Search any
task
Search any
task
