List your company

Spatial decision-making on MiniBehavior size 20

72.96Success RateSpecFlow
2235.2348.4661.69Jun 1, 2026

Evaluation results

MethodLinks
SpecFlowSetting=Flow(image) + AR(Text)2026.0672.96
Gemini-3-FlashSetting=Naive2026.0670.19
DiffThinkerSetting=Flow(image) Only2026.0666.58
GPT-5.1Setting=Naive2026.0666.14
MVoTSetting=Autoregressive2026.0665.23
Qwen3-VL-32BSetting=SFT2026.0648.72
Qwen3-VL-32BSetting=GRPO2026.0644.35
Qwen3-VL-8BSetting=SFT2026.0638.94
Qwen3-VL-8BSetting=GRPO2026.0635.02
Qwen3-VL-32BSetting=Naive2026.0631.63
Qwen3-VL-8BSetting=Naive2026.0623.96