List your company

Puzzle Solving Agentic Task on Sokoban

89.84Success RateL1
73.428877.689481.9586.2106Jun 12, 2026

Evaluation results

MethodLinks
L1Backbone Model=Qwen3-4B, Metric Aggregation Type=Best2026.0689.846.53
L2Backbone Model=Qwen3-4B, Metric Aggregation Type=Best2026.0688.286.44
ReProBackbone Model=Qwen3-4B, Metric Aggregation Type=Best2026.0688.286.61
GiGPOBackbone Model=Qwen3-4B, Metric Aggregation Type=Best2026.0686.726.21
Meta PromptBackbone Model=Qwen3-4B, Metric Aggregation Type=Best2026.0685.165.94
ReProBackbone Model=Qwen3-4B, Metric Aggregation Type=Avg2026.0681.645.1
L2Backbone Model=Qwen3-4B, Metric Aggregation Type=Avg2026.0680.084.79
L1Backbone Model=Qwen3-4B, Metric Aggregation Type=Avg2026.0679.224.56
GiGPOBackbone Model=Qwen3-4B, Metric Aggregation Type=Avg2026.0676.094.21
Meta PromptBackbone Model=Qwen3-4B, Metric Aggregation Type=Avg2026.0674.063.91