OpenMed is a local-first healthcare AI toolkit for clinical entity extraction and PII de-identification that runs entirely on your own device or infrastructure.
OpenMed is an open-source healthcare AI project focused on turning clinical text into structured output. The README says it supports entity extraction, PII de-identification, and a curated catalog of more than 1,000 specialized medical models. It is designed to run locally on CPUs, CUDA systems, and Apple hardware, including native iPhone, iPad, and Mac apps through OpenMedKit.
The project addresses the need to analyze healthcare text without sending sensitive patient data to a cloud vendor. It targets settings where privacy, compliance, offline operation, or air-gapped deployment matter, while still needing specialized medical NLP capabilities such as named entity recognition and PII removal.
Conceptually, OpenMed takes clinical text as input and returns medical entities, detected PII, and de-identified text. The README shows multiple access paths: a Python API, a REST service, batch processing, and native Swift usage through OpenMedKit. It also indicates that model selection can be done by model name or local model path, and that Apple devices can use MLX acceleration with a CoreML fallback path, while non-Apple environments can use the Hugging Face runtime.
It is gaining attention because it combines a strong privacy message with practical developer ergonomics: one-line Python usage, REST deployment, batch processing, and native iOS/macOS support. The repository also emphasizes local-first healthcare AI, offline operation, Apple Silicon acceleration, multilingual support, and a large specialized model registry, which makes it relevant to current interest in on-device AI and sovereign AI workflows.
The README explicitly contrasts OpenMed with cloud medical APIs, positioning it as an on-device or on-premise alternative. Within the project itself, the main related approaches mentioned are its Python package, its REST service, its batch processing workflow, and its native OpenMedKit integration for Apple platforms. The README does not name competing open-source projects directly, so no broader comparison can be stated from the provided material alone.
AI-explained · grounded in each repo's README