Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Spatial Reasoning QA on SAT-Real

88.7Average Accuracy

Gemini-3.0 + 3D Belief

85.16486.0828787.918May 12, 2026
Updated 21d ago

Evaluation Results

MethodLinks
2026.05
88.710073.997.394.175.8
2026.05
86.7957094.189.781.3
2026.05
8695.778.386.594.175.8
2026.05
85.395.778.383.885.384.8