SecureBio HPCT: leaderboard

SecureBio HPCT — 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 April 2027. 111 models tracked. Updated 16 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1Claude Opus 4 [16k]SecureBio run: native-setting=Claude Opus 4 [16k]91.67
2Claude Opus 4.6Max80.72#106
3Kimi K2.679.02#181
4Claude Opus 5Max78.65#20
5Claude Sonnet 5.5Max77.2#19
6GPT-5.6 SolMax76.32#30
7Gemini 4 ArgonHigh74.15#12
8GPT-6 AstraMax68.96#6
9Claude Opus 4.7Max68.33#78
10GPT-5.6 TerraMax68.19#64
11Claude Opus 4.5 [16k]SecureBio run: native-setting=Claude Opus 4.5 [16k]67.91
12GPT-6 SolMax67.77#26
13GPT-5.5xHigh66.79#52
14Claude Sonnet 5Max66.47#111
15GLM-5.2Max65.28#141

No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #10 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=securebio-hpct · How It Works · Data refreshed daily, snapshot 2026-11-16.