Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-image Understanding on Blink
Loading...
61.18
Accuracy
Ours
38.1128
44.1014
50.09
56.0786
Jul 1, 2025
Aug 14, 2025
Sep 27, 2025
Nov 11, 2025
Dec 25, 2025
Feb 7, 2026
Mar 24, 2026
Accuracy
Updated 5d ago
Evaluation Results
Method
Method
Links
Accuracy
Ours
Training=RL-enhanced
2025.07
61.18
Qwen2.5-VL-7B
Model Size=7B
2025.07
58.07
Qwen2.5-VL-7B + SFT
Training=SFT
2025.07
56.56
LLaVA
2026.03
40.4
VISOR
2026.03
39
Feedback
Search any
task
Search any
task