Benchmarks.bio - BioSecBench-Surveillance: leaderboard
Metric: Pass Rate (%). Source: benchmarks.bio. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 5 | 60.8 |
| 2 | Grok 4.6 | 53.5 |
| 3 | Grok 4.5 | 53.3 |
| 4 | GPT-5.6 Sol | 52.5 |
| 5 | GPT-5.6 Terra | 51 |
| 6 | GPT-5.6 Luna | 50.3 |
| 7 | GPT-5.5 | 50.2 |
| 8 | Claude Opus 4.8 | 50.2 |
| 9 | Claude Sonnet 5 | 49.8 |
| 10 | Claude Opus 4.7 | 49.6 |
| 11 | Claude Sonnet 4.6 | 48.6 |
| 12 | Kimi K3 | 48.3 |
| 13 | Gemini 3.5 Flash | 47.1 |
| 14 | Claude Opus 4.6 | 46.7 |
| 15 | Gemini 3.1 Pro (Preview) | 45.3 |
Interactive version: theaggregate.ai/benchmark?slug=benchmarks-bio-biosecbench-surveillance · How It Works · Data refreshed daily, snapshot 2026-09-05.