Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Visual Reasoning on SAT
Loading...
76
Accuracy
Qwen3-VL-32B
8.0568
25.6959
43.335
60.9741
Jun 17, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Qwen3-VL-32B
3-run average=true
2026.06
76
GPT-5
3-run average=true
2026.06
73.3
Qwen3-VL-8B
3-run average=true
2026.06
54
Visual-OPSD
3-run average=true, La...
2026.06
54
ThinkMorph
3-run average=true, La...
2026.06
52.67
InternVL3.5-38B
3-run average=true
2026.06
49.33
Visual-OPSD-Noise
3-run average=true, La...
2026.06
48
Text-only SFT
3-run average=true, La...
2026.06
46.63
InternVL3.5-8B
3-run average=true
2026.06
45.33
BAGEL-7B
3-run average=true
2026.06
44.67
GPT-4o
3-run average=true
2026.06
28
Janus-Pro-7B
3-run average=true
2026.06
22
Chameleon-7B
3-run average=true
2026.06
10.67
Feedback
Search any
task
Search any
task