SecRespond - Detection: leaderboard
Metric: Detection CHK-score (%): the detection (discovery, evidence and root-cause attribution, 0-3 per checkpoint) score achieved over each range checkpoints (averaged over Claude Opus 4.7, Gemini 3.1 Pro and GPT-5.4 Pro as judges), averaged over the 10 SecRespond cyber ranges; agents in the OpenCode harness with reasoning enabled investigate a forensic disk snapshot of a compromised host and write forensic reports and a remediation plan; higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 23 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 (Thinking) | 78.2 |
| 2 | Qwen 3.7 Max (Thinking) | 78 |
| 3 | GLM-5.1 | 76.3 |
| 4 | Claude Sonnet 4.6 (Thinking) | 74.7 |
| 5 | GPT-5.4 Pro (xHigh) | 72.5 |
| 6 | GPT-5.5 | 70.7 |
| 7 | Claude Sonnet 4.5 (Thinking) | 69.7 |
| 8 | Claude Opus 4.5 (Thinking) | 69.5 |
| 9 | Qwen 3.6 Plus (Thinking) | 68.4 |
| 10 | GPT-5.2 Pro | 66.2 |
| 11 | Kimi K2.6 (Thinking) | 65.8 |
| 12 | GLM-5 | 64.9 |
| 13 | Qwen 3.5 Plus (Thinking) | 63.5 |
| 14 | Kimi K2.5 (Thinking) | 62.1 |
| 15 | GPT-5.4 | 60.7 |
Interactive version: theaggregate.ai/benchmark?slug=secrespond-detection · How It Works · Data refreshed daily, snapshot 2026-09-29.