We're excited to announce that we're officially re-launching MedARC! 🚀
MedARC is an open science research collective, aimed to accelerate medical AI via a collaborative science-in-the-open approach. (1/5)
MedReasoner workshop at @CVPR is just starting in Room 110, packed with some excellent speakers in the medical AI space.
Honored to be an invited speaker for the workshop, come check out my talk in an hour (2:20pm)!!
A big focus at Sophont is building foundation models to understand & analyze the brain. This is still a nascent field but we believe such models could eventually help improve diagnosis & treatment of neurodegenerative diseases and mental disorders.
This project is some of our initial efforts in the space. We have trained a family of foundation models on functional MRI (fMRI) neuroimaging data that achieves SOTA performance on a variety of benchmarks. We do so by introducing a novel approach for representing fMRI data called flat maps.
Of course, since it's still an emerging area, there is a lack of systematic benchmarks, which we attempt to fix with our Brainmarks benchmarking suite.
All code and models are completely open-sourced!
As usual, this project was done with the support of our broader @MedARC_AI community. If you're interested in contributing to this line of research, please join the MedARC discord! We are now building multimodal MRI foundation models and participating in the FOMO challenge.
If you are interested in partnering on neuro foundation models, be sure to contact me directly as well!
We're excited to announce we're starting a Journal Club. And our first meeting is scheduled for tomorrow!
@__init_self will present her work, CorText: Brain-Language Fusion Enables Interactive Neural Readout and In-Silico Experimentation
Tomorrow at 10:15am ET, join Discord!
We're excited to announce we're starting a Journal Club. And our first meeting is scheduled for tomorrow!
@__init_self will present her work, CorText: Brain-Language Fusion Enables Interactive Neural Readout and In-Silico Experimentation
Tomorrow at 10:15am ET, join Discord!
Medmarks: A Comprehensive Open-Source LLM Benchmark Suite for Medical Tasks
We (@SophontAI) recently released our medical benchmarking research on arXiv.
"We introduce Medmarks, a fully open-source evaluation suite with 30 benchmarks spanning question answering, information extraction, medical calculations, and open-ended clinical reasoning. We perform a systematic evaluation of 61 models across 71 configurations using verifiable metrics and LLM-as-a-Judge. Our results show that frontier reasoning models (Gemini 3 Pro Preview, GPT-5.1, & GPT-5.2) achieve the highest performance across both benchmarks, most frontier proprietary models are significantly more token efficient than open-weight alternatives, medically fine-tuned models outperform their generalist counterparts, and that models are susceptible to answer-order bias"
We're excited to release Medmarks v1.0 + a technical report!
This is an update to our Medmarks benchmark suite, the largest open-source automated suite for evaluating the medical capabilities of LLMs.
We added 10 benchmarks (20→30) and 15 models (46→61) to the leaderboard!
We're excited to release Medmarks v1.0 + a technical report!
This is an update to our Medmarks benchmark suite, the largest open-source automated suite for evaluating the medical capabilities of LLMs.
We added 10 benchmarks (20→30) and 15 models (46→61) to the leaderboard!
A good discussion about evaluating LLMs in medicine by the head of Health AI at OpenAI, highly recommend reading.
Appreciate the shoutout for Medmarks, our LLM evaluation suite developed at @SophontAI/@MedARC_AI. Glad to see even folks at frontier labs are finding it useful!
This release was only possible by the numerous MedARC volunteers who implemented benchmarks and datasets to evaluate with. Grateful to all those who contributed!
We're releasing Medmarks v0.1, the largest completely open-source automated evaluation suite for assessing the medical capabilities of LLMs!
Developed in our @MedARC_AI community, w/ support from @PrimeIntellect
So far we’ve explored 46 models to figure out the best!
We're releasing Medmarks v0.1, the largest completely open-source automated evaluation suite for assessing the medical capabilities of LLMs!
Developed in our @MedARC_AI community, w/ support from @PrimeIntellect
So far we’ve explored 46 models to figure out the best!
Sophont had a great NeurIPS last week!
We presented our fMRI foundation model research at the Brain and Body Foundation Models workshop (which we also co-sponsored!)
We also held a social with @KindredVentures (and @_CausalLabs, @fal) including a great panel with our CEO Tanishq Abraham, MIT prof Paul Liang, Stanford prof James Zou, moderated by Kanyi Maqubela from Kindred. We discussed the importance of multimodal models for medicine, open-source, open research problems in the space, agents for accelerating scientific discovery, and so much more.
It was great to connect with so many folks interested in medical AI, see you next time!
Sophont had a great NeurIPS last week!
We presented our fMRI foundation model research at the Brain and Body Foundation Models workshop (which we also co-sponsored!)
We also held a social with @KindredVentures (and @_CausalLabs, @fal) including a great panel with our CEO Tanishq Abraham, MIT prof Paul Liang, Stanford prof James Zou, moderated by Kanyi Maqubela from Kindred. We discussed the importance of multimodal models for medicine, open-source, open research problems in the space, agents for accelerating scientific discovery, and so much more.
It was great to connect with so many folks interested in medical AI, see you next time!
Check out OpenMidnight, our SOTA pathology foundation model.
Available on HuggingFace!
This model can be fine-tuned for a variety of usecases:
- classifying cancer type
- segmenting cells and tissue
- predicting genes activity from pathology image
and so much more!
NEW RELEASE:
"How to Train a State-of-the-Art Pathology Foundation Model with $1.6k"
We present OpenMidnight, a our first pathology foundation model, trained on just whole-slide images for only $1.6K!
We open-source model weights and reproducible training code!
Excited to share our latest @SophontAI release 🥳
"How to Train a State-of-the-Art Pathology Foundation Model with $1.6k"
We present OpenMidnight, our first pathology foundation model!
It has SOTA perf. despite being only trained on 12k whole slide images w/ $1.6k compute!
Some personal news: I've joined @SophontAI to help build the next generation of open medical foundation models.
We've relaunched @MedARC_AI, our open science research community. Join us if you want to help advance open medical AI.
And we are hiring.
Sharing our work on training fMRI neuroimaging foundation models, to be presented at FMs for Brain & Body NeurIPS 2025 workshop
Although this was an internal Sophont research project, we are now expanding this at MedARC
If you're interested in contributing, join us in Discord!!
Excited to share our first paper:
Scaling Vision Transformers for Functional MRI with Flat Maps
We introduce a new approach to training fMRI neuroimaging foundation models and demonstrate a strict dataset power scaling law!
Our team is working hard and doing great work, we have some cool research stuff to share this month, excited to release!
Be sure to follow @SophontAI and @MedARC_AI if you want to stay up to date!
At @MedARC_AI we are building a comprehensive suite of medical LLM evals, and we already have tons of volunteers and lots of great progress!
The project started less than a week ago!
Are there other medical LLM evals we should include?