List your company

Vision Reasoning on BabyVision

94.1FD ScoreHuman
5.80428.72751.6574.573May 13, 2026

Evaluation results

MethodLinks
Human2026.0594.194.194.194.1
Gemini 3.1 Pro2026.0546.253.753.943.4
GPT-5.22026.0527.335.254.934.9
CAVE-7BParam=7B, Instruct=false2026.0515.318.725.514.3
InternVL3.5Param=8B, Instruct=true2026.0514.121.919.610.4
Qwen3-VLParam=8B, Instruct=true2026.0513.514.311.79.09
CAVE-3BParam=3B, Instruct=false2026.0513.517.623.510.4
CAVE Belief-OnlyParam=7B, Instruct=false2026.0512.918.715.79.1
MiniCPM-V 4.5Param=8B, Instruct=false2026.0512.319.819.69.09
VLAA-ThinkerParam=7B, Instruct=false2026.0512.320.919.611.7
DeepEyesParam=7B, Instruct=false2026.0511.518.722.110.9
ThinkLite-VLParam=7B, Instruct=false2026.051117.621.611.7
Qwen2.5-VLParam=7B, Instruct=true2026.0510.413.113.77.79
Random2026.0510.310.310.310.3
R1-OneVisionParam=7B, Instruct=false2026.059.216.523.511.7