SecRespond - Planning: leaderboard
SecRespond (Alibaba and HKUST, 2026) remediation planning: agents in the OpenCode harness propose fixes for the intrusions, persistence and risks on 10 compromised cloud-host cyber ranges. The score is the planning checkpoint score (correct and complete remediation) as a percent of the maximum, averaged over ranges and three LLM judges; higher is better.
Metric: Planning CHK-score (%): the remediation-planning (correct and complete fix, 0-2 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: Not forecast. 23 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 (Thinking) | 61.2 | #250 (Claude Sonnet 4.5) |
| 2 | GLM-5.1 | 59.2 | #179 |
| 3 | Claude Sonnet 4.6 (Thinking) | 59.1 | #161 (Claude Sonnet 4.6) |
| 4 | Claude Opus 4.6 (Thinking) | 58 | #102 (Claude Opus 4.6) |
| 5 | Claude Opus 4.5 (Thinking) | 56.4 | #156 (Claude Opus 4.5) |
| 6 | Qwen 3.7 Max (Thinking) | 54.3 | #145 (Qwen 3.7 Max) |
| 7 | Kimi K2.5 (Thinking) | 50.7 | #256 (Kimi K2.5) |
| 8 | Qwen 3.6 Plus (Thinking) | 48.3 | #222 (Qwen 3.6 Plus) |
| 9 | GLM-5 | 45.9 | #260 |
| 10 | Qwen 3.5 Plus (Thinking) | 44 | #232 (Qwen 3.5 Plus) |
| 11 | DeepSeek V3.2 (Thinking) | 43.6 | #406 (DeepSeek V3.2) |
| 12 | Kimi K2.6 (Thinking) | 43 | #178 (Kimi K2.6) |
| 13 | GPT-5.4 Pro (xHigh) | 40.7 | #85 (GPT-5.4 Pro) |
| 14 | MiniMax-M2.7 | 40 | #518 |
| 15 | GPT-5.2 Pro | 37.1 | #124 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=secrespond-planning · How It Works · Data refreshed daily, snapshot 2026-11-13.