List your company

Spatial Reasoning on Robospatial (ood)

70.2AccuracyThinking with Blueprint
47.63253.49159.3565.209Jan 5, 2026

Evaluation results

MethodLinks
Thinking with BlueprintModel Size=7B, Reasoning Mode=Thinking with Blueprint2026.0170.2
VigoRLModel Size=7B, Reasoning Mode=Thinking with Coordinates2026.0167.9
Qwen2.5-VLModel Size=7B, Reasoning Mode=No2026.0166.7
Qwen2.5-VLModel Size=7B, Reasoning Mode=Naive CoT2026.0166.3
gpt-5-ThinkingModel Size=-, Reasoning Mode=Naive CoT2026.0165.4
gpt-4oModel Size=-, Reasoning Mode=Naive CoT2026.0163.6
gpt-4oModel Size=-, Reasoning Mode=No2026.0160.1
ViLASRModel Size=7B, Reasoning Mode=Thinking with Images2026.0148.5