List your company

Sequential Planning on Visual Spatial Planning (VSP) FrozenLake

100VSP FrozenLake Success Rate (Level 3)Gemini-3-Flash
30.3248.4166.584.59Dec 30, 2025

Evaluation results

MethodLinks
Gemini-3-FlashSetting=N/A2025.12100100100998398
DiffThinker++Setting=Flow Matching2025.1210010010098100100
GPT-5Setting=N/A2025.12997067433629
Qwen3-VL-8BSetting=SFT2025.12999698969286
Qwen3-VL-32BSetting=GRPO2025.12999095697358
DiffThinkerSetting=Flow Matching2025.12991009899100100
Qwen3-VL-32BSetting=SFT2025.129699981009990
Qwen3-VL-8BSetting=GRPO2025.12917070313424
Qwen3-VL-32BSetting=N/A2025.12755147252326
Qwen3-VL-8BSetting=N/A2025.12644633211221
Qwen-Image-Edit-2511Setting=N/A2025.12505544162323
Qwen-Image-Edit-2509Setting=N/A2025.1233362212117