Xent Games: leaderboard
Interactive game-agent leaderboard where LLMs play games in LLM-generated contexts with LLM-enforced rules and LLM-scored outcomes.
Metric: Overall Score (self-reported). Source: benchmarklist.com. Status: saturation imminent. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 65.86 |
| 2 | Grok 4 | 63.22 |
| 3 | GPT-5 | 62.77 |
| 4 | DeepSeek Reasoner | 62.67 |
| 5 | Gemini 2.5 Flash | 59.08 |
| 6 | Claude Opus 4.1 (20250805) | 58.65 |
| 7 | Claude Opus 4 (20250514) | 58.35 |
| 8 | GPT-5 Mini | 49.22 |
| 9 | Claude Sonnet 4 (20250514) | 48.45 |
| 10 | Kimi K2 0905 | 42.89 |
| 11 | DeepSeek V3 Chat | 35.48 |
| 12 | GPT-5 Nano | 23.27 |
Interactive version: theaggregate.ai/benchmark?slug=xent-games · How It Works · Data refreshed daily, snapshot 2026-09-05.