poker_ai and pluribus
About poker_ai
fedden/poker_ai
🤖 An Open Source Texas Hold'em AI
Implements Counterfactual Regret Minimization (MCCFR) through iterative self-play to learn Nash equilibrium strategies, using information set clustering to compress the massive game tree. Provides a complete pipeline: card abstraction via lossy/lossless compression, strategy training with resumable checkpoints, and interactive play via CLI or terminal UI against trained agents.
About pluribus
zanussbaum/pluribus
An attempt at a Python implementation of Pluribus, a No-Limits Hold'em Poker Bot
This project helps you explore and understand strategies for playing poker variants, specifically Kuhn Poker and Leduc Hold'em. It takes in the rules of these simplified poker games and outputs optimal playing strategies based on Monte Carlo Counterfactual Regret Minimization and Depth Limited Search. This is useful for researchers in game theory or AI, or anyone interested in developing advanced poker AI.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work