List your company
ResearchFormerly WizWand

Multi-agent game strategic reasoning on G game v0.8A (test)

84LLM Only PerformanceK-R
16.433.9551.569.05May 8, 2026
Updated 5mo ago

Evaluation results

MethodLinks
K-RBase model=GPT-4o, Warm-up phase=5 episodes, Contextual history=Provided2026.058454482445174245
SOMBase model=GPT-4o, Warm-up phase=5 episodes, SCM structure=Fixed2026.058061593947196453
CoTBase model=GPT-4o, Warm-up phase=5 episodes, Contextual history=Provided2026.05681654283293635
ReflexionBase model=GPT-4o, Warm-up phase=5 episodes, Contextual history=Provided2026.056410402026235434
ToTBase model=GPT-4o, Warm-up phase=5 episodes, Contextual history=Provided2026.054618221222112622
LLM onlyBase model=GPT-4o, Warm-up phase=5 episodes, Contextual history=Provided2026.0519412210378