List your company

Spatial Reasoning on SpatialScore 1.0 (test)

86.6Overall ScoreHuman-level
25.957641.701357.44573.1887May 22, 2025

Evaluation results

MethodLinks
Human-levelModel Category=Baselines2025.0586.696.8789.7282.3392.1585.1881.6478.9694.4686.8984.19
Gemini-3-ProModel Scale=Proprietary, Rank=12025.0560.1270.462.9958.1562.4460.0264.249.571.1251.3759.93
GPT-5Model Scale=Proprietary, Rank=22025.0558.1378.0857.5955.1354.0459.2267.3945.0157.1150.962.13
Qwen3-VL-235B-A22BModel Scale=30B-78B, Rank=12025.0556.6357.2765.1954.0452.4759.970.0150.469.442.849.63
VST-7B-RLModel Scale=7B-14B, Rank=12025.0552.4741.3954.9256.85055.4864.5648.6563.3742.845.59
VST-3B-RLModel Scale=1B-4B, Rank=12025.0548.0424.6152.3851.7543.7254.1661.8443.0362.8938.9547.43
VST-3B-SFTModel Scale=1B-4B, Rank=22025.0547.3725.7348.4553.1145.7453.0261.4142.3560.4835.3550.74
Spatial-MLLM-4BModel Scale=1B-4B, Rank=32025.0542.5933.5647.6134.843.2756.1657.6840.4936.3928.4153.31
Blind GPT-5 (Text-only)Model Category=Baselines2025.0530.6218.7920.3429.3640.8131.0545.3424.225.5432.0126.47
Chance-level (Random)Model Category=Baselines2025.0528.2923.7122.822.8831.8430.2938.5924.0625.0628.7928.68