OpenMed equips clinical engineering and data science teams with open-source LLMs, biomedical NER, and evaluation tooling to ship HIPAA-aware workflows faster—without vendor lock-in.
Production-ready state-of-the-art LLMs for healthcare and clinical AI across biomedical domains — chemicals, diseases, genes/proteins, species, anatomy, oncology, and more. Advanced healthcare AI models trained on clinical and biomedical data.
Explore our comprehensive collection of state-of-the-art models for clinical NLP, biomedical NER, and healthcare applications.
Specialized model for Chemical Entity Recognition - Identifies chemical compounds and substances in biomedical literature.
Specialized model for Disease Entity Recognition - Disease entities from the BC5CDR dataset.
Specialized model for Gene Entity Recognition - Gene-related entities.
Specialized model for Cancer Genetics - Cancer-related genetic entities.
Specialized model for Biomedical Entity Recognition - Proteins, DNA, RNA, cell lines, and cell types.
Specialized model for Chemical Entity Recognition - Chemical entities from the BC5CDR dataset.
Production-ready healthcare AI models deployed on AWS SageMaker with enterprise-grade security and scalability.
Scalable cloud infrastructure with automatic provisioning and load balancing
Optimized inference endpoints with sub-100ms latency for real-time applications
HIPAA-compliant infrastructure with end-to-end encryption and audit trails
Open-source gene entity recognition tuned on BC2GM for precision genomics workflows. Deploy secure inference endpoints in under five minutes using the managed marketplace package.
Identifies organism mentions in clinical narratives to accelerate biosurveillance, antimicrobial stewardship and microbiome research pipelines.
Hands-on notebooks for deployment, scaling, monitoring and cost optimization across every OpenMed marketplace model.
Peer-reviewed research advancing the state of healthcare AI with rigorous scientific methodology and reproducible results.
Named-entity recognition (NER) is fundamental to extracting structured information from the >80% of healthcare data that resides in unstructured clinical notes and biomedical literature. Despite recent advances with large language models, achieving state-of-the-art performance across diverse entity types while maintaining computational efficiency remains a significant challenge. We introduce OpenMed NER, a suite of open-source, domain-adapted transformer models that combine lightweight domain-adaptive pre-training (DAPT) with parameter-efficient Low-Rank Adaptation (LoRA). Our approach performs cost-effective DAPT on a 350k-passage corpus compiled from ethically sourced, publicly available research repositories and de-identified clinical notes (PubMed, arXiv, and MIMIC-III) using DeBERTa-v3, PubMedBERT, and BioELECTRA backbones. This is followed by task-specific fine-tuning with LoRA, which updates less than 1.5% of model parameters. We evaluate our models on 12 established biomedical NER benchmarks spanning chemicals, diseases, genes, and species. OpenMed NER achieves new state-of-the-art micro-F1 scores on 10 of these 12 datasets, with substantial gains across diverse entity types. Our models advance the state-of-the-art on foundational disease and chemical benchmarks (e.g., BC5CDR-Disease, +2.70 pp), while delivering even larger improvements of over 5.3 and 9.7 percentage points on more specialized gene and clinical cell line corpora. This work demonstrates that strategically adapted open-source models can surpass closed-source solutions. This performance is achieved with remarkable efficiency: training completes in under 12 hours on a single GPU with a low carbon footprint (< 1.2 kg CO2e), producing permissively licensed, open-source checkpoints designed to help practitioners facilitate compliance with emerging data protection and AI regulations, such as the EU AI Act.
Answers to the most common questions from clinical innovation, ML, and compliance teams evaluating OpenMed for production NLP and decision support workloads.
Each model ships with benchmark results across 12 biomedical NER datasets, guardrail evaluations, and Apache-2.0 licensing, so teams can satisfy procurement reviews and accelerate go-live timelines.
Use the AWS Marketplace model packages and JumpStart notebooks for guided SageMaker deployment, or pull the Docker images and Python package to host inside your own VPC or on hospital infrastructure.
Yes. Models are trained on de-identified, ethically sourced corpora, support private deployment, and integrate audit logging, encryption, and access controls when run on SageMaker or self-managed environments.
Absolutely. Lightweight LoRA adapters, curated tokenizers, and starter notebooks are provided so you can extend entity coverage to local ontologies or multi-lingual records while keeping compute requirements modest.
Founded by Maziyar Panahi, OpenMed is a community-driven, non-profit effort to democratize state-of-the-art LLMs for healthcare and make powerful clinical AI freely available. All healthcare AI models are released under Apache-2.0 with practical demos and deployment recipes for medical and clinical applications.
For interviews, speaking, or partnerships, reach out and we'll get back quickly.