Wiz Cyber Model Arena (ADK React) - Cloud Security: leaderboard
Wiz Cyber Model Arena scores AI agents on offensive security challenges in four categories (code vulnerabilities, API security, Websec.fr CTFs, cloud security). This board is the percentage of cloud security challenges solved on the first attempt (pass@1) by models run in the ADK React harness; higher is better.
Metric: Cloud Security challenges solved, pass@1 (%). Source: www.wiz.io. Saturation forecast: Around July 2027. 21 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Argon | 82.6 | #9 |
| 2 | Claude Opus 5.5 | 66.7 | #3 |
| 3 | GPT-5.6 Sol | 62.3 | #22 |
| 4 | Claude Opus 5 | 60.9 | #16 |
| 5 | Gemini 3.6 Flash | 56.5 | #57 |
| 6 | Gemini 3.7 Flash | 56.5 | #24 |
| 7 | GPT-5.6 Terra | 55.1 | #40 |
| 8 | DeepSeek V4 Pro | 53.6 | #112 |
| 9 | Claude Opus 4.8 | 52.2 | #39 |
| 10 | DeepSeek V4.1 Flash (1017) | 52.2 | #27 |
| 11 | DeepSeek V4.1 Flash | 50.7 | #37 |
| 12 | Claude Haiku 5.5 | 50.7 | #43 |
| 13 | Kimi K3 | 49.3 | #35 |
| 14 | Claude Sonnet 4.5 | 47.8 | #148 |
| 15 | GLM-5.3 | 47.8 | #44 |
No result here: #6 GPT-6 Astra, #7 GPT-6.1 Sol, #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=wiz-cyber-model-arena-adk-react-cloud-security · How It Works · Data refreshed daily, snapshot 2026-11-05.