Home / AI Trending / maziyarpanahi/openmed

openmed — What is it?

OpenMed is a local-first healthcare AI that transforms clinical text into structured insights, enabling on-device processing without patient data leaving the network.

⭐ 4,255 Stars 🍴 497 Forks Python Apache-2.0 Author: maziyarpanahi
Source: per README View on GitHub →

Why it matters

OpenMed is gaining attention due to its unique focus on local-first processing, addressing privacy concerns in healthcare AI. It fills the gap by providing a comprehensive suite of medical NLP tools that run entirely on-device, with support for Apple Silicon and MLX acceleration, making it stand out in the healthcare AI space.

Source: Synthesis of README and project traits

Core Features

Local-first processing

All processing occurs on the device, ensuring patient data never leaves the network, addressing privacy concerns in healthcare AI.

Source: per README
1,000+ specialized medical models

A vast collection of models for entity extraction, PII de-identification, and other medical NLP tasks, covering a wide range of clinical and biomedical domains.

Source: per README
Multilingual support

Support for 12 languages, making it versatile for global healthcare applications.

Source: per README
Apple Silicon and MLX acceleration

Optimized for Apple Silicon with MLX acceleration, providing significant performance improvements for on-device processing.

Source: per README

Architecture

The architecture is modular, with clear separation of concerns. It includes a Python API for model interaction, a REST service for server-side processing, and a batch processing module. The codebase is structured with a focus on scalability and maintainability, utilizing design patterns like dependency injection and separation of concerns. Data flows through the system in a controlled manner, with clear interfaces for each component.

Source: Code tree + dependency files

Project Knowledge Graph

Knowledge graph: project (center) + core features (inner hexagons) + key dependencies (outer chips) transformers huggingface-hub accelerate tokenizers mlx Local-first processingLocal-first process… 1,000+ specialized medical models1,000+ specialized… Multilingual support Apple Silicon and MLX accelerationApple Silicon and M… openmed Project Core feature Key dependency

Center: project; inner ring: core feature modules; outer ring: key dependencies. Auto-generated from core_features and tech_stack.key_deps.

Tech Stack

LanguagePythonFrameworkFastAPI for REST service, PyTorch for machine learning models, and OpenMedKit for iOS/macOS integration.
Key dependencies
transformershuggingface-hubacceleratetokenizersmlxcoremltools
Infrastructure / Deployment
Docker for containerization, potentially running on various cloud platforms or on-premises infrastructure.
Source: Dependency files + code tree

Quick Start

pip install 'openmed[hf]' pip install 'openmed[hf,service]' pip install 'openmed[mlx]' # Python API from openmed import analyze_text analyze_text('Patient received 75mg clopidogrel for NSTEMI.', model_name='pharma_detection_superclinical') # REST service uvicorn openmed.service.app:app --host 0.0.0.0 --port 8080 # Batch from openmed import BatchProcessor p = BatchProcessor(model_name='disease_detection_superclinical', group_entities=True) p.process_texts([...])
Source: README Installation/Quick Start

Use Cases

OpenMed is suitable for healthcare organizations, researchers, and developers in the medical AI field. It is useful in scenarios such as clinical documentation analysis, patient data de-identification, and medical research. Specific problems it solves include extracting medical entities from clinical text, de-identifying patient information, and enabling on-device AI processing for privacy and efficiency.

Source: README

Strengths & Limitations

Strengths

  • Strength 1: Strong focus on privacy and local-first processing, addressing critical concerns in healthcare AI.
  • Strength 2: Comprehensive suite of medical NLP tools with a wide range of specialized models.
  • Strength 3: Support for Apple Silicon and MLX acceleration for improved performance.

Limitations

  • Limitation 1: Limited information on performance metrics and scalability on large datasets.
  • Limitation 2: Dependency on Python and specific frameworks may limit accessibility for some users.
Source: Synthesis of README, code structure and dependencies

Latest Release

v1.5.5 (2026-06-08): Added batch PII extraction/de-identification, service lifecycle, and README refresh. v1.5.2 (2026-05-27): Security and MLX conversion hardening release. v1.5.1 (2026-05-21): Patch release for OpenMed 1.5.1. v1.5.0 (2026-05-19): Added Arabic, Japanese, and Turkish support. v1.4.0 (2026-05-11): Multilingual Privacy Filter release.

Source: GitHub Releases

Verdict

OpenMed is a promising project for healthcare AI applications, particularly for those requiring strong privacy guarantees and on-device processing capabilities. It is well-suited for teams and individuals working in the healthcare industry or related fields, with a focus on medical NLP and AI.

Frequently Asked Questions

What is openmed?

OpenMed is a local-first healthcare AI that transforms clinical text into structured insights, enabling on-device processing without patient data leaving the network.

What are the main features of openmed?

openmed's core features include: Local-first processing, 1,000+ specialized medical models, Multilingual support, Apple Silicon and MLX acceleration.

Why is openmed trending?

OpenMed is gaining attention due to its unique focus on local-first processing, addressing privacy concerns in healthcare AI.

What is openmed used for?

OpenMed is suitable for healthcare organizations, researchers, and developers in the medical AI field.

Transparency Notice
This page is auto-generated by AI (a large language model) from the following public materials: GitHub README, code tree, dependency files and release notes. Analyzed at: 2026-06-12 18:34. Quality score: 85/100.

Data sources: README, GitHub API, dependency files