Diplomacy: Betrayal Tendency — leaderboard
Metric: Betrayal Rate (%). Source: goodstartlabs.com. 17 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 100 |
| 2 | Hermes 4 405B | 91.7 |
| 3 | GLM-4.6 | 85 |
| 4 | Kimi K2 | 76.7 |
| 5 | Claude Opus 4.5 | 66.7 |
| 6 | Gemini 2.5 Flash | 51.7 |
| 7 | O3 | 41.7 |
| 8 | DeepSeek R1 | 40 |
| 9 | Claude Sonnet 4.5 | 31.7 |
| 10 | Qwen 3 Max | 21.7 |
| 11 | Llama 4 Maverick | 13.3 |
| 12 | Gemini 3 Pro | 11.7 |
| 13 | GPT-5 | 0 |
| 14 | GPT-5 Mini | 0 |
| 15 | GPT-OSS-120B | 0 |
Interactive version: theaggregate.ai/benchmark?slug=diplomacy-betrayal-tendency · How the rankings work · Data refreshed daily, snapshot 2026-07-22.