Atlas

Benchmarks

← All benchmarks

WCB (SecureBio)

Science

SecureBio's World-Class Biology benchmark, 96 text-only open-ended questions graded against an expert rubric by a model grader. Evaluated in Inspect with zero-shot prompting, usually over 10 runs. SecureBio doesn't state how refusals enter accuracy. Claude Opus 5.5 prints 54.5% with 74.4% 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 568.8SecureBio Biology Benchmark Dashboard · SecureBio
Opus 4.866.7SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5.6 Sol64.4SecureBio Biology Benchmark Dashboard · SecureBio
GPT-6 Astra62.6SecureBio Biology Benchmark Dashboard · SecureBio
Opus 4.761.1SecureBio Biology Benchmark Dashboard · SecureBio
Gemini 4 Argon61.1SecureBio Biology Benchmark Dashboard · SecureBio
GPT-6 Sol60.5SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5.6 Terra59.8SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5.559.7SecureBio Biology Benchmark Dashboard · SecureBio
Gemini 3.7 Flash58.0SecureBio Biology Benchmark Dashboard · SecureBio
Grok 4.757.6SecureBio Biology Benchmark Dashboard · SecureBio
Gemini 3.8 Flash57.1SecureBio Biology Benchmark Dashboard · SecureBio
Gemini 3.5 Flash56.7SecureBio Biology Benchmark Dashboard · SecureBio
Opus 4.656.5SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5.5 Pro56.5SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5.2 Pro56.3SecureBio Biology Benchmark Dashboard · SecureBio
Sonnet 4.655.4SecureBio Biology Benchmark Dashboard · SecureBio
Kimi K355.0SecureBio Biology Benchmark Dashboard · SecureBio
GPT-554.6SecureBio Biology Benchmark Dashboard · SecureBio
Claude Opus 5.554.5SecureBio Biology Benchmark Dashboard · SecureBio
Gemini 3 Pro Preview54.2SecureBio Biology Benchmark Dashboard · SecureBio
Kimi K2.654.1SecureBio Biology Benchmark Dashboard · SecureBio
o353.8SecureBio Biology Benchmark Dashboard · SecureBio
Gemini 3.1 Pro Preview53.3SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5.153.2SecureBio Biology Benchmark Dashboard · SecureBio
Gemini 3.6 Flash52.2SecureBio Biology Benchmark Dashboard · SecureBio
Kimi K3.1 Pro51.5SecureBio Biology Benchmark Dashboard · SecureBio
Qwen3.8 2.4T A95B50.5SecureBio Biology Benchmark Dashboard · SecureBio
Grok 4.650.3SecureBio Biology Benchmark Dashboard · SecureBio
Qwen3.7-Max (source unspecified)50.2SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5 Mini49.2SecureBio Biology Benchmark Dashboard · SecureBio
Inkling49.1SecureBio Biology Benchmark Dashboard · SecureBio
Sonnet 549.0SecureBio Biology Benchmark Dashboard · SecureBio
Beam 1.0 501B A23B48.8SecureBio Biology Benchmark Dashboard · SecureBio
Opus 4.547.9SecureBio Biology Benchmark Dashboard · SecureBio
o4-mini47.5SecureBio Biology Benchmark Dashboard · SecureBio
GLM-5.247.0SecureBio Biology Benchmark Dashboard · SecureBio
GLM-5.346.6SecureBio Biology Benchmark Dashboard · SecureBio
Muse Glimmer 30B46.6SecureBio Biology Benchmark Dashboard · SecureBio
GPT-5.446.4SecureBio Biology Benchmark Dashboard · SecureBio
Loading Atlas data…