SecureBio VCT v2: leaderboard
SecureBio VCT v2 — Native SecureBio benchmark panel; exact named population. Accuracy includes native refusal/failure conventions and source-reported runs. Raw fractions converted only according to this panel: ABLEFull already uses percent, baseline-only and other capability panels use fractions. Metric: Accuracy (%). Source: https://securebio.org/benchmarks/details/
Metric: Accuracy (%). Source: securebio.org. Saturation forecast: Around 2029. 27 models tracked. Updated 16 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Claude Opus 5.5 | Max | 60.54 | #5 |
| 2 | GPT-6 Astra | Max | 58 | #6 |
| 3 | Gemini 4 Argon | High | 57.35 | #12 |
| 4 | Claude Sonnet 5.5 | Max | 52.18 | #19 |
| 5 | GPT-Rosalind-5.5 | xHigh | 50.29 | #46 |
| 6 | GPT-5.6 Sol | Max | 49.45 | #30 |
| 7 | GPT-6 Sol | Max | 49.35 | #26 |
| 8 | Gemini 3.7 Flash | High | 48.63 | #34 |
| 9 | Gemini 3.5 Flash | High | 46.51 | #100 |
| 10 | Kimi K3.1 Pro | Max | 46.24 | #29 |
| 11 | Claude Opus 4.8 | Max | 45.87 | #58 |
| 12 | GPT-5.5 Pro | xHigh | 45.34 | #35 |
| 13 | GPT-5.5 | xHigh | 45.27 | #52 |
| 14 | Gemini 3.8 Flash | High | 45.23 | #36 |
| 15 | Gemini 3.1 Pro (Preview) | High | 43.64 | #93 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #10 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=securebio-vct-v2 · How It Works · Data refreshed daily, snapshot 2026-11-16.