#2929th of 50 in Data and scraping for AIFor science, health and research
openmed
Clinical NER and PII de-identification SDK that runs locally on CPU, GPU, Apple, mobile
- Stars
- 5.8k
- License
- Apache-2.0
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- Python
834 on 21 Apr 20265,799 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
OpenMed is a Python SDK with a registry of clinical NER and PII models. It extracts diseases, drugs, anatomy and genes from medical text and de-identifies it by mask, replace, hash or date shift. It runs on CPU, CUDA, Apple MLX, Android via ONNX Runtime Mobile and browsers via Transformers.js, and also offers a REST service, an MCP server and agent skills.
Who it is for: Healthcare developers de-identifying and extracting entities from clinical text
Strengths
- Same extract_pii and deidentify API across PyTorch, MLX, ONNX, Swift, Kotlin and Transformers.js
- Runs offline from a local model directory without contacting the Hugging Face Hub
- PII support for 42 language codes, 35 of them model-backed
- Ships REST endpoints, an MCP server and agent skills
Weaknesses
- Only the SDK source is Apache-2.0; model and dataset licenses vary
- SDK use alone does not establish HIPAA compliance; expert review is required
- Clinical brief demo uses synthetic test doubles, not validated models
- Mobile, MLX and browser paths need per-model exported artifacts
What it needs
- GPU optional
- Docker + Compose
- Needs Hugging Face Hub (model download), PyTorch, MLX (optional, Apple Silicon), ONNX Runtime (optional)
- Models: Hugging Face PyTorch, MLX, ONNX, CoreML, Transformers.js
- port 8080
Quick answers
Is openmed open source?
Yes. Its code is on GitHub under the Apache-2.0 license.
Can you run openmed with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Does openmed need a GPU?
No. It runs on a CPU and can use a GPU when you have one.
Also in Data and scraping for AI
See all 50| Rank | Project | Score |
|---|---|---|
| 1 | doclingPython library that converts PDFs, Office files and more into structured text | 92 out of 100 |
| 2 | mineruParses PDFs, images and Office files into Markdown and structured JSON | 91 out of 100 |
| 3 | opendataloader-pdfPDF parser that outputs Markdown, JSON with bounding boxes, and Tagged PDF | 90 out of 100 |
| #4 | agent-reachCLI that gives coding agents read access to login-gated and Chinese web platforms | 88 out of 100 |
| #5 | markitdownPython library and CLI that converts office files, PDFs and audio to Markdown | 87 out of 100 |