List your company

Spatial Reasoning on MINDCUBE 1.0 (test)

47.62Overall ScoreDeepSeek-VL2-Small
21.817628.516335.21541.9137Jun 26, 2025

Evaluation results

MethodLinks
DeepSeek-VL2-SmallModel Category=Open-Weight Multi Image Model2025.0647.623750.3826.91
GPT-4o-2024-08-06Model Category=Proprietary Model2025.0647.5993.3334.1741.63
LLaVA-Onevision-7BModel Category=Open-Weight Multi Image Model2025.0647.4336.4548.4244.09
Gemini-1.5-pro-002Model Category=Proprietary Model2025.0647.0585.525.9538.4
Gemma-2-9B-itModel Category=Open-Weight Multi Image Model2025.0646.6738.3948.3834.63
mPLUG-Owl3-7B-241101Model Category=Open-Weight Multi Image Model2025.0644.8537.8447.1126.91
Claude-3.5-Sonnet-20240620Model Category=Proprietary Model2025.0644.7548.4244.2147.62
LLaVA-Video-Qwen-7BModel Category=Open-Weight Multi Image Model2025.0641.9635.7143.5530.12
Mantis-8B (SigLip)Model Category=Open-Weight Multi Image Model2025.0641.0537.6540.2350.99
InternVL2-8BModel Category=Open-Weight Multi Image Model2025.0637.52642.0336
RoboBrainModel Category=Spatial Model2025.0637.3835.838.2829.53
Space-QwenModel Category=Spatial Model2025.0633.2838.0233.7126.32
Qwen2-VL-2B-InstructModel Category=Open-Weight Multi Image Model2025.0633.2137.3733.2630.34
Random (frequency)Model Category=Baseline2025.0633.0238.332.6635.79
Random (chance)Model Category=Baseline2025.0632.3536.3632.2930.66
Spatial-MLLMModel Category=Spatial Model2025.0632.0638.3920.9232.82
LongVA-7BModel Category=Open-Weight Multi Image Model2025.0629.4635.8929.5524.88
Qwen2-VL-7B-InstructModel Category=Open-Weight Multi Image Model2025.0629.2638.7629.521.35
SpaceMantisModel Category=Spatial Model2025.0622.8137.6521.2629.32