SecRespond - Baseline Risk Planning: leaderboard
Metric: Planning CAP-score (%) on the Baseline Risk (BAS) 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: Around 2029. 23 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 (Thinking) | 70.8 |
| 2 | Claude Sonnet 4.6 (Thinking) | 69.8 |
| 3 | Claude Opus 4.6 (Thinking) | 67.6 |
| 4 | GLM-5.1 | 65.7 |
| 5 | Qwen 3.7 Max (Thinking) | 63.7 |
| 6 | Claude Opus 4.5 (Thinking) | 63.5 |
| 7 | Qwen 3.6 Plus (Thinking) | 62.6 |
| 8 | GLM-5 | 54.1 |
| 9 | Kimi K2.5 (Thinking) | 52.9 |
| 10 | GPT-5.4 Pro (xHigh) | 51.9 |
| 11 | MiniMax-M2.7 | 51.2 |
| 12 | DeepSeek V3.2 (Thinking) | 50 |
| 13 | Qwen 3.5 Plus (Thinking) | 49 |
| 14 | Kimi K2.6 (Thinking) | 48.6 |
| 15 | GPT-5.5 | 46.8 |
Interactive version: theaggregate.ai/benchmark?slug=secrespond-baseline-risk-planning · How It Works · Data refreshed daily, snapshot 2026-09-29.