Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-modal Understanding on SEED-ALL
Loading...
61.6
Accuracy
LLaVA-1.5 Vanilla 13B
56.9512
58.1581
59.365
60.5719
Aug 25, 2025
Accuracy
Relative Performance
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Relative Performance
LLaVA-1.5 Vanilla 13B
Tokens Retained=576, C...
2025.08
61.6
100
VisionZip
Tokens Retained=192, C...
2025.08
61.2
99.4
MMTok
Tokens Retained=192, C...
2025.08
61.17
99.3
DivPrune
Tokens Retained=192, C...
2025.08
60.83
98.8
MMTok
Tokens Retained=128, C...
2025.08
60.11
97.6
VisionZip
Tokens Retained=128, C...
2025.08
59.74
97
MMTok
Tokens Retained=64, Co...
2025.08
59.51
96.6
DivPrune
Tokens Retained=128, C...
2025.08
59.49
96.6
DivPrune
Tokens Retained=64, Co...
2025.08
57.7
93.7
VisionZip
Tokens Retained=64, Co...
2025.08
57.13
92.7
Feedback
Search any
task
Search any
task
