Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on MathVision (Score, #F)
Loading...
41.1
Score
AVIS
21.652
26.701
31.75
36.799
Jun 10, 2026
Score
#F
Updated 1mo ago
Evaluation Results
Method
Method
Links
Score
#F
AVIS
Checkpoint=Vision-R1 [...
2026.06
41.1
0.78
Fixed
Checkpoint=Vision-R1 [...
2026.06
40.8
1.2
Baseline
Checkpoint=Vision-R1 [...
2026.06
39.5
1
AVIS
Checkpoint=VL-Rethinke...
2026.06
37.3
0.82
Fixed
Checkpoint=VL-Rethinke...
2026.06
36.5
1.25
Baseline
Checkpoint=VL-Rethinke...
2026.06
33.2
1
Fixed
Checkpoint=OpenVLThink...
2026.06
28.6
1.05
AVIS
Checkpoint=OpenVLThink...
2026.06
28.4
0.77
Baseline
Checkpoint=OpenVLThink...
2026.06
27.9
1
Fixed
Checkpoint=Qwen 2.5 VL...
2026.06
25.4
1.04
AVIS
Checkpoint=Qwen 2.5 VL...
2026.06
25
0.78
Baseline
Checkpoint=Qwen 2.5 VL...
2026.06
22.4
1
Feedback
Search any
task
Search any
task
