List your company

Spatial decision-making on MiniBehavior size 12

91.27Success RateSpecFlow
44.480456.627768.77580.9223Jun 1, 2026

Evaluation results

MethodLinks
SpecFlowSetting=Flow(image) + AR(Text)2026.0691.27
Gemini-3-FlashSetting=Naive2026.0690.36
DiffThinkerSetting=Flow(image) Only2026.0688.31
GPT-5.1Setting=Naive2026.0687.45
MVoTSetting=Autoregressive2026.0686.78
Qwen3-VL-32BSetting=SFT2026.0672.16
Qwen3-VL-32BSetting=GRPO2026.0669.04
Qwen3-VL-8BSetting=SFT2026.0663.18
Qwen3-VL-8BSetting=GRPO2026.0659.41
Qwen3-VL-32BSetting=Naive2026.0656.71
Qwen3-VL-8BSetting=Naive2026.0646.28