Frontier AI for human biology, starting with the brain.

All-in-human foundation models to transform neuroscience drug development.

Our Mission

Over the last decade, we have built one of the largest proprietary collections of multimodal patient brain data. We’ve used that data to train foundation models that can reason about what’s happening inside a patient’s brain, and predict how it will change with treatment. We’re now using that platform to usher in a new era of precision neurology, where we can predict which targets to drug, and in which patients.

Run smaller, faster, lower-risk clinical trials.

Enriching a Parkinson’s trial for L-DOPA responders with vBx-1.0 lifts the responder fraction from 52% to 69% — shrinking the required enrollment by 43% at fixed power.

0 20 40 60 80 100 50 100 150
Statistical power (%)
Trial size (# patients)
vBx-1.0 requires 43% fewer patients

01

Human Data

Multi-modal patient data directly from the brain.

The brain is the molecular ground truth for neurological disease, but it’s one of the hardest datasets to collect. Over a decade, we’ve built one of the world’s largest proprietary multi-modal neuroscience patient datasets of its kind: deep molecular profiling on 12,000+ brains, across 6,000 patients including paired proteomic, genomic, and clinical data. Plus a physical inventory of 900+ frozen brain tissue samples spanning the major neurodegenerative diseases.

02

Foundation Models

Building world models of patient biology.

We are building the first AI system designed to reason about neurological disease at the patient level. The output is a virtual biopsy of the brain — a reconstructed picture of what’s happening inside a patient’s brain, generated from a routine blood draw.

03

Applications

From better targets to better trials.

Our platform is solving the hardest problems in drug discovery — which targets to drug, in which patients — to increase the probability of clinical success. Drug developers can use our platform to power target discovery, target characterization, biomarker identification, and patient stratification — the full set of questions needed to design programs that work in humans.

Key Metrics

$-46.2B
In partnerships with top-20 pharma
-8692+
Novel drug targets surfaced
-2763%
Preclinical validation rate
-71
AI-discovered drugs to reach clinical candidates

Our Research

News· June 17, 2026
STAT News: “Verge Labs’ new AI model solves patient stratification problems for neurology clinical trials”
Verge’s CEO, Alice Zhang, and Head of Product & Engineering, Emily Ripka, spoke to STAT News about the release of vBx-1.0, the company’s breakthrough foundation model for precision neurology.

News· June 16, 2026
Verge Labs Unveils vBx: A Breakthrough Foundation Model for Precision Neurology
The new ‘virtual biopsy’ model predicts brain activity, therapy response, and side effects in patients with up to 3x greater accuracy than previous state-of-the-art models.

Report· June 16, 2026
Introducing vBx-1.0
Verge’s multi-modal foundation model for precision neurology.