InteractiveBench Texas Hold'em: leaderboard
Texas Hold'em interactive-game domain: 5,000 no-limit hands across 10 tables.
Metric: Average winnings per hand (chips; blinds 50/100, six-player tables). Source: www.interactivebench.org. Source: www.interactivebench.org. Saturation forecast: Around April 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash | 31.8 | #99 |
| 2 | Grok 4.1 Fast | 27.9 | #217 |
| 3 | GPT-5 Mini | 22.2 | #186 |
| 4 | DeepSeek V3.2 | -23.2 | #205 |
| 5 | Kimi K2 (Thinking) | -28.3 | #243 (Kimi K2) |
| 6 | Qwen 3 Max | -30.4 | #208 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=interactivebench-texas-hold-em · How It Works · Data refreshed daily, snapshot 2026-10-20.