PokéLLMon
P
Pokéllmon
Overview :
POKéLLMON is the first LLM-based agent to achieve human-level performance in a tactical combat game. It leverages three key strategies: 1) Contextual reinforcement learning, which uses text-based feedback from battles to iteratively optimize its generation strategy; 2) Knowledge-enhanced generation, which utilizes external knowledge to combat hallucinations and enable the agent to act accurately and timely; 3) Self-consistent action generation, which mitigates panicking switching behavior when facing strong opponents and avoiding combat. Battles against human players demonstrate POKéLLMON's human-level fighting ability and strategy, achieving a 49% win rate in tournament matches and a 56% win rate in invitational matches. Furthermore, it reveals vulnerabilities in human players' exhaustion strategies and deceptive tactics.
Target Users :
Used for online Pokémon battles with human players.
Total Visits: 29.7M
Top Region: US(17.94%)
Website Views : 51.9K
Use Cases
Engage in real-time Pokémon battles with other players on the PokéLLMon website.
Experience PokéLLMon's human-level combat performance with your friends.
Learn excellent tactical combat strategies through PokéLLMon.
Features
Contextual Reinforcement Learning
Knowledge-Enhanced Generation
Self-Consistent Action Generation
Online Battle with Human Players
Human-Level Combat Performance and Strategy
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase