poke-env and meloetta
These are complementary tools where poke-env provides the higher-level RL training framework that meloetta's lower-level battle client and library could theoretically integrate with, though poke-env's maturity and adoption suggest it serves as the de facto standard interface for this use case.
About poke-env
hsahovic/poke-env
A python interface for training Reinforcement Learning bots to battle on pokemon showdown
This is a Python library that lets you build and test AI players for the online game Pokemon Showdown. You can feed in your bot's logic or a reinforcement learning algorithm, and it will control Pokémon battles, allowing you to observe how your AI performs. It's designed for developers interested in creating and evaluating AI for Pokémon battles.
About meloetta
spktrm/meloetta
A Pokémon Battle Client and Reinforcement Learning Library for Interacting with Pokémon Showdown written in Python
Scores updated daily from GitHub, PyPI, and npm data. How scores work