![[2509.22387] SpinGPT: A Large-Language-Model Approach to Playing Poker Correctly](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2509.22387] SpinGPT: A Large-Language-Model Approach to Playing Poker Correctly
Summary
SpinGPT introduces a novel approach using large language models to enhance poker strategies, particularly in the Spin & Go format, achieving significant accuracy in decision-making.
Why It Matters
The development of SpinGPT represents a significant advancement in applying AI to complex, multi-player games like poker. By leveraging large language models, it addresses the limitations of traditional algorithms, potentially transforming strategies in competitive gaming and AI research.
Key Takeaways
- SpinGPT is the first large language model designed for the Spin & Go poker format.
- It combines supervised fine-tuning and reinforcement learning to improve decision-making.
- Achieves 78% accuracy in matching solver actions and performs well against established poker bots.
Computer Science > Machine Learning arXiv:2509.22387 (cs) [Submitted on 26 Sep 2025 (v1), last revised 23 Feb 2026 (this version, v2)] Title:SpinGPT: A Large-Language-Model Approach to Playing Poker Correctly Authors:Narada Maugin, Tristan Cazenave View a PDF of the paper titled SpinGPT: A Large-Language-Model Approach to Playing Poker Correctly, by Narada Maugin and 1 other authors View PDF Abstract:The Counterfactual Regret Minimization (CFR) algorithm and its variants have enabled the development of pokerbots capable of beating the best human players in heads-up (1v1) cash games and competing with them in six-player formats. However, CFR's computational complexity rises exponentially with the number of players. Furthermore, in games with three or more players, following Nash equilibrium no longer guarantees a non-losing outcome. These limitations, along with others, significantly restrict the applicability of CFR to the most popular formats: tournaments. Motivated by the recent success of Large Language Models (LLM) in chess and Diplomacy, we present SpinGPT, the first LLM tailored to Spin & Go, a popular three-player online poker format. SpinGPT is trained in two stages: (1) Supervised Fine-Tuning on 320k high-stakes expert decisions; (2) Reinforcement Learning on 270k solver-generated hands. Our results show that SpinGPT matches the solver's actions in 78% of decisions (tolerant accuracy). With a simple deep-stack heuristic, it achieves 13.4 +/- 12.9 BB/100 versus Slu...
