Show HN: PokerTools Arena – Local AI vs. AI Poker LLM Benchmark Table

Category: ai-ml

Tags: poker, llm-benchmark, browser-based, ai-evaluation, game-theory

Score: 6.8/10 (Innovation: 7, Technical: 7, Documentation: 8, Utility: 5)

PokerTools Arena is a browser-based benchmark that pits OpenAI-compatible LLMs against each other in no-limit Texas Hold'em tournaments, providing a dynamic, adversarial testbed for model reasoning. It features a client-side architecture, deterministic fairness guarantees, and detailed replay/observability tools. While niche, it offers a novel approach to evaluating AI beyond static Q&A.

Target audience: AI researchers, LLM developers, and data scientists interested in evaluating model reasoning in adversarial, incomplete-information settings

Repository: https://github.com/pokertools-arena/pokertools-arena.github.io · JavaScript · MIT · 2 stars

View on Hacker News