SecRespond - Vulnerability Risk Planning: leaderboard
SecRespond (Alibaba and HKUST, 2026) incident-response agents in the OpenCode harness, scored on the checkpoints mapped to the vulnerability risk capability dimension across 10 compromised cloud-host cyber ranges. The score is the remediation-planning score achieved as a percent of the maximum over those checkpoints, averaged over three LLM judges; higher is better.
Metric: Planning CAP-score (%) on the Vulnerability Risk (VUL) capability dimension: the remediation-planning (correct and complete fix, 0-2 per checkpoint) score achieved over all checkpoints mapped to the dimension across the 10 SecRespond cyber ranges (averaged over three LLM judges); agents in the OpenCode harness with reasoning enabled; higher is better. Source: arxiv.org. Saturation forecast: Not forecast. 23 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.7 (Thinking) | 72.6 | #79 (Claude Opus 4.7) |
| 2 | GLM-5.1 | 71.1 | #179 |
| 3 | Claude Sonnet 4.6 (Thinking) | 67.8 | #161 (Claude Sonnet 4.6) |
| 4 | Claude Sonnet 4.5 (Thinking) | 66.7 | #250 (Claude Sonnet 4.5) |
| 5 | Claude Opus 4.6 (Thinking) | 63.3 | #102 (Claude Opus 4.6) |
| 6 | Kimi K2.5 (Thinking) | 60 | #256 (Kimi K2.5) |
| 7 | GLM-5 | 58.8 | #260 |
| 8 | DeepSeek V3.2 (Thinking) | 54.5 | #406 (DeepSeek V3.2) |
| 9 | Claude Opus 4.5 (Thinking) | 54.4 | #156 (Claude Opus 4.5) |
| 10 | Qwen 3.6 Plus (Thinking) | 52.3 | #222 (Qwen 3.6 Plus) |
| 11 | Qwen 3.7 Max (Thinking) | 47.8 | #145 (Qwen 3.7 Max) |
| 12 | GPT-5.4 Pro (xHigh) | 45.6 | #85 (GPT-5.4 Pro) |
| 13 | Qwen 3.5 Plus (Thinking) | 44.4 | #232 (Qwen 3.5 Plus) |
| 14 | Gemini 3 Flash | 43.3 | #164 |
| 15 | Kimi K2.6 (Thinking) | 41.1 | #178 (Kimi K2.6) |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=secrespond-vulnerability-risk-planning · How It Works · Data refreshed daily, snapshot 2026-11-13.