# PokerSkill: LLMs Can Play Expert-Level Poker without Training or Solvers

Source: [PokerSkill: LLMs Can Play Expert-Level Poker without Training or Solvers](<https://www.sota2.com/research/paper/6a191de3dbd7150506f82c06>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-30T05:00:11.986Z

Authors: Boning Li, Baoxiang Wang, Longbo Huang

Published: 2026-05-28T15:38:33.000Z

## Abstract

Poker is a landmark challenge for artificial intelligence. The dominant approach relies on equilibrium solvers built on counterfactual regret minimization, requiring millions of core-hours of training. Large Language Models (LLMs) possess extensive poker knowledge but perform far below solver-based agents when asked to play directly. Traditional rule-based poker agents are interpretable and training-free, but their strategic ceiling remains far below equilibrium play. We introduce \\textbf{PokerSkill}, a training-free and solver-free framework that bridges this gap by using detailed rule-based poker skills as a structured action-grounding interface for LLMs. A deterministic context engine analyzes the current state and retrieves only the relevant fragments from a layered skill library, which is entirely designed by human poker experts, constraining the LLM's choice to reasonable actions. Against GTOWizard, a state-of-the-art GTO benchmark, GPT-5.5 XHigh with PokerSkill achieves $-57 \\pm 21$ mbb/hand, Claude Opus 4.6 achieves $-80 \\pm 29$ mbb/hand and Claude Opus 4.7 achieves $-87\\pm 64$ mbb/hand, reducing losses by 49--61\\% compared to default-prompt baselines and outperforming the strong bot Slumbot. Our key finding is that rule-based skills alone do not constitute a strong strategy, and LLMs alone cannot play well, but their combination yields an agent that requires neither training nor solver access yet competes with systems built on millions of core-hours of computation. To our knowledge, this is the first demonstration of an LLM achieving competitive performance in a complex imperfect-information game without game-specific training or solver queries. Code is available at https://github.com/lbn187/PokerSkill.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Heads-Up No-Limit Hold'em (HUNL) on GTOWizard benchmark](<https://www.sota2.com/research/sota/heads-up-no-limit-hold-em-hunl-on-gtowizard-benchmark.md>): Heads-Up No-Limit Hold'em (HUNL); GTOWizard benchmark; mbb/h: 61.

- [Heads-up No-Limit Texas Hold'em on GTOWizard HUNL](<https://www.sota2.com/research/sota/heads-up-no-limit-texas-hold-em-on-gtowizard-hunl.md>): Heads-Up No-Limit Texas Hold'em; GTOWizard HUNL; BB/h: 132.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2605.30094>)

[PDF](<https://arxiv.org/pdf/2605.30094.pdf>)