THE FRONTIER MEDICAL AI LAB / EST. 2025

Building towards openmedical superintelligence forclinicians

Multimodal foundation models for medicine. Open weights. Evaluated publicly. Built with the community.

Explore Available Models Partner with Us

§ 01OUR VISION

Building a universal
medical foundation model.

Working toward a unified, multimodal understanding of patient state.

Input patient data

Unify every modality

📋

Clinical Notes

Visits, history, symptoms

🧪

Labs

Blood, urine, chemistry

🩻

Imaging

X-ray, CT, MRI

🔬

Pathology

H&E, IHC, slides

💓

Signals

ECG, wearables

🧬

Genomics

DNA, RNA, proteomics

💊

Medications

Treatments, response

👤

Demographics

Lifestyle, context

Universal medical AI

Learn functional biology of patient state

Clinical applications

Predict and intervene

Model disease trajectory

Predict how a patient's state evolves over time and their risk of adverse outcomes.

Predict intervention response

Identify interventions that can move a patient toward a healthier state.

Accelerate biomedical research

Generate clinically grounded hypotheses from multiscale, longitudinal patient representations.

§ 02RESEARCH

Pushing the frontier with open models and rigorous benchmarking

Explore our available models, benchmark leaderboards, and active open challenges.

FEATURED · PATHOLOGY CHALLENGE

nanopath

A challenge to train the best computational pathology foundation model in 1 hour on 1 gpu

PATHOLOGY · MODEL

OpenMidnight

Open-source pathology foundation model for biomarker discovery and predicting clinical outcomes

FMRI · MODEL

CortexMAE

Open-source functional MRI foundation model to improve signal quality and predict clinical biomarkers of neurodegenerative and psychiatric disorders.

Clinical text · BENCHMARK

Medmarks

Open medical LLM benchmark suite evaluating clinically relevant capabilities, with a live leaderboard

FMRI · BENCHMARK

Brainmarks

An open fMRI benchmark suite for predicting brain state and clinical diagnoses

§ 03COMMUNITY

Open science
community.

MedARC is the world's largest open medical AI research community — founded by Sophont's CEO and home to thousands of researchers, clinicians, and engineers.

Community

MedARC

The world's largest open medical AI research community.

What happens in MedARC

→Open collaborative research projects

→Regular journal clubs and reading groups

→Shared compute and dataset access

→Direct access to Sophont researchers

4,000+

Community members

Published papers

Join MedARC

Journal Club

Join the next discussion.

We invite researchers to present their work, then open the room for discussion with the community. Add the series to your calendar, or sign up below for email notifications.

Add to calendar

GCal event iCal Public calendar

§ 04SOLUTIONS

Open models drive
clinical applications.

We develop commercial products with pharma, hospitals, and medtech teams, translating our open-source models into clinically relevant applications.

IN DEVELOPMENT

Coming Soon

Our first commercial product is in active development with clinical, biopharma, and life sciences partners. We'll announce more when it's ready — with rigorous validation.

Stay tuned

§ 05PUBLICATIONS

Publications.

Selected peer-reviewed papers, workshop contributions, and technical reports from the Sophont and MedARC research collective.

Jun 2026

Compress-Distill: Reasoning Trace Compression for Efficient Knowledge Distillation

M. Griot, P.S. Scotti, T.M. Abraham

arXiv ↗

May 2026

Medmarks: A comprehensive open-source LLM benchmark suite for medical tasks

B. Warner, R.S. Grandhi, M. Kieffer, ..., C. Lane, P.S. Scotti, T.M. Abraham

ICML Workshop ↗

May 2026

Scaling Vision Transformers for Functional MRI with Flat Maps

C. Lane, M. Tripathy, L.K. Murali, R.S. Grandhi, ..., D.Z. Kaplan, B. Warner, T.M. Abraham, P.S. Scotti

ICML ↗

Dec 2025

MIRAGE: Robust multi-modal architectures translate fMRI-to-image models from vision to mental imagery

R. Kneeland, C. Torrico, T. Chen, J.A. Ojeda, S. Khanna, J. Xu, P.S. Scotti, T. Naselaris

DBM 2025 Findings ↗

Nov 2025

OpenMidnight: How to Train a State-of-the-Art Pathology Foundation Model with $1.6k

D. Kaplan, R.S. Grandhi, C. Lane, B. Warner, T.M. Abraham, P. Scotti

Sophont TR ↗

Jun 2025

CheXAgent: A vision-language foundation model to enhance efficiency of chest X-ray interpretation

Z. Chen, M. Varma, J. Xu, ..., C. Bluethgen, T.M. Abraham, ..., J.-B. Delbrouck, A.S. Chaudhari, C.P. Langlotz

arXiv ↗

Dec 2023

Reconstructing the mind’s eye: fMRI-to-image with contrastive learning

P.S. Scotti, A. Banerjee, J. Goode, ..., K.A. Norman, T.M. Abraham

NeurIPS Spotlight ↗

§ 06TEAM

Built by researchers
from the field.

A small team of medical AI researchers with roots in academic hospitals, open-source ML, and NeuroAI.

Tanishq Mathew Abraham, Ph.D.

CEO & Co-Founder

Former Research Director at Stability AI; founded MedARC, the world’s largest online medical AI research community. Graduated at 19 with a Ph.D. in Biomedical Engineering at University of California, Davis.

UC DavisStability AI