Wiz Cyber Model Arena (Claude Code) - Code Vulnerabilities: 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 code vulnerability challenges solved on the first attempt (pass@1) by models run in the Claude Code harness; higher is better.
Metric: Code Vulnerabilities challenges solved, pass@1 (%). Source: www.wiz.io. Saturation forecast: Around December 2027. 25 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Argon | 59.9 | #9 |
| 2 | Claude Opus 5.5 | 58.5 | #3 |
| 3 | GPT-6 Astra | 56.6 | #6 |
| 4 | GPT-6.1 Sol | 56.6 | #7 |
| 5 | Claude Opus 5 | 55.2 | #16 |
| 6 | GPT-6 Sol | 50.8 | #19 |
| 7 | Claude Haiku 5.5 | 46.6 | #43 |
| 8 | Claude Opus 4.8 | 44 | #39 |
| 9 | GPT-6 Luna | 43.4 | #61 |
| 10 | Gemini 3.6 Flash | 39.4 | #57 |
| 11 | GPT-5.6 Terra | 38.9 | #40 |
| 12 | DeepSeek V4 Pro | 37.9 | #112 |
| 13 | Gemini 3.5 Flash | 37.7 | #66 |
| 14 | DeepSeek V4.1 Flash (1017) | 37.1 | #27 |
| 15 | Claude Sonnet 5 | 36.3 | #65 |
No result here: #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=wiz-cyber-model-arena-claude-code-code-vulnerabilities · How It Works · Data refreshed daily, snapshot 2026-11-05.