Atlas

Benchmarks

← All benchmarks

MBCT (SecureBio)

Science

SecureBio's Molecular Biology Capabilities Test, 200 text-only multiple-response questions on tacit knowledge in routine molecular biology methods. Evaluated in Inspect with zero-shot prompting, usually over 10 runs. SecureBio doesn't state how refusals enter accuracy. Claude Opus 5.5 prints 68.7% with 58.6% refused, more than it could score if refusals counted as wrong, so accuracy appears to cover answered questions only and each model's denominator differs. These scores aren't comparable with other runs of the test, which is why this is a diagnostic.

Benchmark source: Securebio

Top models (higher is better)

ModelScoreSource
Claude Opus 5.568.7SecureBio Biology Benchmark Dashboard · SecureBio
GPT-6 Astra63.9SecureBio Biology Benchmark Dashboard · SecureBio
Gemini 4 Argon63.2SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5.6 Sol61.1SecureBio Biology Benchmark Dashboard · SecureBio
GPT-6 Sol61.1SecureBio Biology Benchmark Dashboard · SecureBio
Kimi K3.1 Pro60.7SecureBio Biology Benchmark Dashboard · SecureBio
Claude Opus 560.2SecureBio Biology Benchmark Dashboard · SecureBio
Kimi K2.659.8SecureBio Biology Benchmark Dashboard · SecureBio
Sonnet 559.2SecureBio Biology Benchmark Dashboard · SecureBio
Opus 459.1SecureBio Biology Benchmark Dashboard · SecureBio
Kimi K358.4SecureBio Biology Benchmark Dashboard · SecureBio
Qwen3.8 2.4T A95B57.8SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5.557.1SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5.5 Pro57.0SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5.6 Terra56.9SecureBio Biology Benchmark Dashboard · SecureBio
Opus 4.656.7SecureBio Biology Benchmark Dashboard · SecureBio
Opus 4.856.4SecureBio Biology Benchmark Dashboard · SecureBio
Opus 4.756.0SecureBio Biology Benchmark Dashboard · SecureBio
Gemini 3 Flash Preview55.6SecureBio Biology Benchmark Dashboard · SecureBio
GLM-5.255.5SecureBio Biology Benchmark Dashboard · SecureBio
Opus 4.155.1SecureBio Biology Benchmark Dashboard · SecureBio
Gemini 3.7 Flash55.0SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5.254.8SecureBio Biology Benchmark Dashboard · SecureBio
Gemini 3.6 Flash54.3SecureBio Biology Benchmark Dashboard · SecureBio
Sonnet 4.554.1SecureBio Biology Benchmark Dashboard · SecureBio
Qwen3.8 27B53.0SecureBio Biology Benchmark Dashboard · SecureBio
Opus 4.552.9SecureBio Biology Benchmark Dashboard · SecureBio
Qwen3.5 397B A17B52.5SecureBio Biology Benchmark Dashboard · SecureBio
Gemini 2.5 Pro52.2SecureBio Biology Benchmark Dashboard · SecureBio
Qwen3.7-Max (source unspecified)51.8SecureBio Biology Benchmark Dashboard · SecureBio
Grok 451.7SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5 Mini51.5SecureBio Biology Benchmark Dashboard · SecureBio
Inkling51.2SecureBio Biology Benchmark Dashboard · SecureBio
o1 (2024-12-17)51.1SecureBio Biology Benchmark Dashboard · SecureBio
GPT-550.8SecureBio Biology Benchmark Dashboard · SecureBio
Grok 4.650.4SecureBio Biology Benchmark Dashboard · SecureBio
GLM-5.349.8SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5 Pro49.6SecureBio Biology Benchmark Dashboard · SecureBio
Beam 1.0 501B A23B49.2SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5.149.2SecureBio Biology Benchmark Dashboard · SecureBio
Loading Atlas data…