CodeClash - Poker: leaderboard

CodeClash arena where LLMs iteratively develop bots to play no-limit Texas Hold'em poker, competing head-to-head across multi-round tournaments.

Metric: ELO. Source: codeclash.ai. Saturation forecast: Around February 2027. 8 models tracked.

Top models

#ModelScoreOverall rank
1GPT-51599#98
2GPT-5 Mini1428#186
3O31278#124
4Claude Sonnet 4.51256#142
5Claude Sonnet 41234#199
6Gemini 2.5 Pro961#152
7Grok Code Fast 1898#337

No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=codeclash-poker · How It Works · Data refreshed daily, snapshot 2026-10-20.