BioSecBench-Surveillance V1 (102 tasks)
Science
BioSecBench-Surveillance V1 expanded 102-evaluation release. Macro-mean pass rate over gradeable genomic surveillance tasks; distinct from the earlier 100-task release and V2 clean reruns.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5 | 51.5 |
| GPT-6 Astra | 48.5 |
| Grok 4.6 | 48.0 |
| Gemini 3.8 Flash | 44.1 |
| Gemini 3.7 Flash | 41.5 |