ResearchBenchmarksPlanning on Commonly solved planning tasks F5-3 vs lmFollow0.33Runtime RatioDeepSeek0.3260.3530.380.407Aug 19, 2025Evaluation ResultsMethodMethodLinksRuntime RatioRuntime Success RatePlan Length RatioPlan Length Equality RateDeepSeekConfiguration=F5-3 (be...Configuration=F5-3 (best run), N=7852025.080.33981.4334QwenConfiguration=F5-3 (be...Configuration=F5-3 (best run), N=7242025.080.36981.6532LlamaConfiguration=F5-3 (be...Configuration=F5-3 (best run), N=5122025.080.37981.7739GPTConfiguration=F5-3 (be...Configuration=F5-3 (best run), N=5362025.080.43972.1928