Benchmarks
Benchmarks and evaluations of AI models across the biosecurity-relevant capabilities that matter, from advancing biological threats to strengthening biodefense. We refresh them over time to prevent saturation.
View the leaderboardWhy now
AI is learning biology fast.
Safety has to scale with it.
Latch builds the measurement layer for biosecurity: the benchmarks, audits, and red-teaming that tell you whether your models meaningfully raise biological risk, and by how much.
Audits
Independent, pre-deployment assessments of the biological risk a frontier model poses, run to feed directly into your responsible scaling policy.
Request an auditRed-teaming
Targeted adversarial testing of a model’s bio-capabilities: pushing on the exact edges a benchmark surfaces.
Agentic KYC
Autonomous identity and legitimacy checks on applicants, so labs can clear managed-access requests in hours instead of weeks.
Our Team
Our team has built biosurveillance across 27 U.S. sites protecting 13M+ people
AI biosecurity benchmarks adopted by frontier AI labs, and pioneered machine learning methods for predicting viral evolution.
Selected papersYear
Paper · 01 / 072026
Deep untargeted wastewater metagenomic sequencing from sewersheds across the United States
medRxiv
A wastewater surveillance initiative capturing 1,206 samples collected between December 2023 and December 2025 from 27 sites across nine states, covering 13 million people. Deep untargeted sequencing enabled detection of SARS-CoV-2, influenza, and emerging pathogens including avian influenza H5N1 — representing 67% of all untargeted wastewater sequencing data currently on the NCBI Sequence Read Archive.