Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
General Multimodal Understanding on Aggregated Benchmarks
Loading...
71
Average Score
Qwen2.5-VL-3B
58.208
61.529
64.85
68.171
May 13, 2026
Average Score
Updated 20d ago
Evaluation Results
Method
Method
Links
Average Score
Qwen2.5-VL-3B
KV Size=256
2026.05
71
Cambrian-4B + Gaze attention
KV Size=288+4
2026.05
69.6
Cambrian-4B + Gaze attention
KV Size=144+4
2026.05
69
Qwen2.5-VL-3B + InfiniPot-V
KV Size=128
2026.05
68.8
LLaVA-OV-7B
KV Size=196
2026.05
68.7
Cambrian-4B
KV Size=576
2026.05
68.6
Cambrian-4B + Gaze attention
KV Size=72+4
2026.05
68.1
Cambrian-4B + TokenCompre. + Gaze attention
KV Size=54+4
2026.05
67.4
Cambrian-4B + TokenCompre.
KV Size=144
2026.05
66.5
Cambrian-4B + TokenCompre. + Gaze attention
KV Size=36+4
2026.05
66.4
LLaVA-OV-7B + HERMES
KV Size=100
2026.05
64.7
Qwen2.5-VL-3B + InfiniPot-V
KV Size=50
2026.05
61.4
LLaVA-OV-7B + HERMES
KV Size=40
2026.05
58.7
Feedback
Search any
task
Search any
task