CWE-bench v2 Public Set Pass@4
Cybersecurity · 2026-11-02
Share of the 100 public CWE-bench v2 tasks solved in at least one of four rollouts.
Benchmark source: Cwe Bench
Top models (higher is better)
| Model | Score | Source |
|---|---|---|
| Gemini 4 Argon | 81.0 | CWE-bench v2 leaderboard (snapshot 2026-11-03) · Collinear AI |
| Muse Wonder | 79.0 | CWE-bench v2 leaderboard (snapshot 2026-11-03) · Collinear AI |
| Claude Opus 5.5 | 78.0 | CWE-bench v2 leaderboard (snapshot 2026-11-03) · Collinear AI |
| Grok 4.7 | 74.0 | CWE-bench v2 leaderboard (snapshot 2026-11-03) · Collinear AI |
| DeepSeek V4.1 Flash 1017 | 72.0 | CWE-bench v2 leaderboard (snapshot 2026-11-03) · Collinear AI |
| GPT-6 Astra | 70.0 | CWE-bench v2 leaderboard (snapshot 2026-11-03) · Collinear AI |
| Hy4 Preview | 68.0 | CWE-bench v2 leaderboard (snapshot 2026-11-03) · Collinear AI |
| GLM-5.3 | 65.0 | CWE-bench v2 leaderboard (snapshot 2026-11-03) · Collinear AI |
| Kimi K3.1 Pro | 63.0 | CWE-bench v2 leaderboard (snapshot 2026-11-03) · Collinear AI |
| Qwen3.8 Max (0902) | 62.0 | CWE-bench v2 leaderboard (snapshot 2026-11-03) · Collinear AI |
| Mistral Large 4 Preview | 59.0 | CWE-bench v2 leaderboard (snapshot 2026-11-03) · Collinear AI |