Wiz Cyber Model Arena (ADK React) - API 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 API security challenges solved on the first attempt (pass@1) by models run in the ADK React harness; higher is better.
Metric: API Security challenges solved, pass@1 (%). Source: www.wiz.io. Saturation forecast: Around April 2027. 21 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Argon | 85.8 | #9 |
| 2 | DeepSeek V4.1 Flash (1017) | 82.7 | #27 |
| 3 | Claude Opus 5 | 81.5 | #16 |
| 4 | DeepSeek V4.1 Flash | 80.9 | #37 |
| 5 | Gemini 3.6 Flash | 80.2 | #57 |
| 6 | Gemini 3.7 Flash | 80.2 | #24 |
| 7 | GPT-5.6 Sol | 78.4 | #22 |
| 8 | GLM-5.3 Flash | 78.4 | #77 |
| 9 | MiMo-V2.6-Pro | 77.8 | #48 |
| 10 | DeepSeek V4 Pro | 77.2 | #112 |
| 11 | Claude Sonnet 5 | 77.2 | #65 |
| 12 | GPT-5.6 Terra | 77.2 | #40 |
| 13 | Claude Opus 4.8 | 76.5 | #39 |
| 14 | Claude Haiku 5.5 | 76.5 | #43 |
| 15 | Claude Opus 5.5 | 75.9 | #3 |
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-api-security · How It Works · Data refreshed daily, snapshot 2026-11-05.