Wiz Cyber Model Arena (ADK React) - Websec.fr CTFs: 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 Websec.fr web capture-the-flag challenges solved on the first attempt (pass@1) by models run in the ADK React harness; higher is better.
Metric: Websec.fr CTFs challenges solved, pass@1 (%). Source: www.wiz.io. Saturation forecast: Around February 2027. 21 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.6 Sol | 85.1 | #22 |
| 2 | Gemini 3.7 Flash | 85.1 | #24 |
| 3 | Gemini 3.6 Flash | 81.6 | #57 |
| 4 | Gemini 3.5 Flash | 75.9 | #66 |
| 5 | GPT-5.6 Terra | 75.9 | #40 |
| 6 | Gemini 4 Argon | 74.7 | #9 |
| 7 | Grok 4.6 | 67.8 | #29 |
| 8 | Claude Opus 5 | 60.9 | #16 |
| 9 | DeepSeek V4.1 Flash (1017) | 60.9 | #27 |
| 10 | Kimi K3 | 59.8 | #35 |
| 11 | Claude Opus 5.5 | 57.5 | #3 |
| 12 | DeepSeek V4 Pro | 52.9 | #112 |
| 13 | GLM-5.3 | 51.7 | #44 |
| 14 | Claude Sonnet 4.5 | 48.3 | #148 |
| 15 | Claude Opus 4.8 | 44.8 | #39 |
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-websec-fr-ctfs · How It Works · Data refreshed daily, snapshot 2026-11-05.