CyberArena - Expected Contest Score: leaderboard
CyberArena:12models in their developer coding harness or OpenCode;40challenge reference set. Conditional posterior mean attack,defense and service-availability score, averaged over opponents/challenges; not observed win rate. Adaptive game selection and native harnesses; requested reasoning effort is not specified. Posterior Expected Score (%).
Metric: Posterior Expected Score (%). Source: arxiv.org. Saturation forecast: Around May 2027. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.6 Sol | 78.6 | #19 |
| 2 | Claude Opus 5 | 73.8 | #12 |
| 3 | GLM-5.3 | 68.5 | #37 |
| 4 | Gemini 3.7 Flash | 67.1 | #22 |
| 5 | Grok 4.6 | 65.5 | #26 |
| 6 | Claude Opus 4.8 | 65 | #36 |
| 7 | Qwen 3.8 Max | 62.3 | #31 |
| 8 | Gemini 3.5 Flash | 60.6 | #61 |
| 9 | DeepSeek V4 Pro (0813) | 59.5 | #52 |
| 10 | MiniMax-M3 | 59.1 | #142 |
| 11 | Kimi K3 | 57 | #30 |
| 12 | Hy4 preview | 56.3 | #39 |
No result here: #4 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=cyberarena-expected-contest-score · How It Works · Data refreshed daily, snapshot 2026-10-22.