List your company
ResearchFormerly WizWand

Long-horizon tasks on Minecraft Diamond

17.36Success Rate (SR)EvoAgent
-0.69443.99288.6813.3672Feb 9, 2025
Updated 6mo ago

Evaluation results

MethodLinks
EvoAgent2025.0217.3626.83
Optimus-12025.029.37.31
Jarvis-12025.027.690.07
DreamerV32025.024.733.69
LS-Imagine2025.024.364.19
PPO2025.0200
GPT-4V2025.0200