Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-skill Compositional Reasoning on MM-Vet
Loading...
32.8
Accuracy
SPARE
22.296
25.023
27.75
30.477
Jun 17, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
SPARE
Backbone=LLaVA-1.5-7B,...
2026.06
32.8
CDPruner
Backbone=LLaVA-1.5-7B,...
2026.06
32
SPARE
Backbone=LLaVA-1.5-7B,...
2026.06
31.6
SPARE
Backbone=LLaVA-1.5-7B,...
2026.06
31.4
LLaVA-1.5-7B
Backbone=LLaVA-1.5-7B,...
2026.06
31.3
CDPruner
Backbone=LLaVA-1.5-7B,...
2026.06
29.8
SparseVLM
Backbone=LLaVA-1.5-7B,...
2026.06
28.9
DART
Backbone=LLaVA-1.5-7B,...
2026.06
28.7
CDPruner
Backbone=LLaVA-1.5-7B,...
2026.06
27.9
PruMerge+
Backbone=LLaVA-1.5-7B,...
2026.06
27.7
DART
Backbone=LLaVA-1.5-7B,...
2026.06
26.6
PruMerge+
Backbone=LLaVA-1.5-7B,...
2026.06
26.3
SparseVLM
Backbone=LLaVA-1.5-7B,...
2026.06
24.3
DART
Backbone=LLaVA-1.5-7B,...
2026.06
22.7
Feedback
Search any
task
Search any
task
