SecureBio ABLE Full: leaderboard
SecureBio ABLE Full — 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 February 2027. 20 models tracked. Updated 16 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | High | 84.62 | #93 |
| 2 | GPT-5.5 Pro | xHigh | 83.57 | #35 |
| 3 | Kimi K2.5 | 82.51 | #256 | |
| 4 | Kimi K3 | Max | 78.43 | #56 |
| 5 | GLM-5.2 | Max | 76.68 | #141 |
| 6 | Gemini 3 Pro (Preview) | High | 74.71 | #113 |
| 7 | Mistral Large 4 | High | 65.31 | #137 |
| 8 | Nemotron 3 Ultra | 62.52 | #428 | |
| 9 | Inkling | xHigh | 59.94 | #281 |
| 10 | Inkling Small | xHigh | 57.5 | #299 |
| 11 | GPT-OSS-120B | 51.71 | #703 | |
| 12 | Beam 1.0 | Max | 50.51 | #206 |
| 13 | Nemotron 3 Super | 49.06 | #775 | |
| 14 | Kimi K2 (Thinking) | 48.12 | ||
| 15 | Qwen 3 235B A22B | Thinking | 45.2 | #628 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=securebio-able-full · How It Works · Data refreshed daily, snapshot 2026-11-16.