Your Data. Your Model. Your Hardware.

Turn clinical text into structured, de-identified insight, with nothing uploaded.
OpenMed extracts biomedical entities and removes 55+ PHI types entirely on the hardware you control, so your data never leaves the device. The same 2,000+ open models run from a phone to a GPU server, fully offline: iOS, iPadOS, and Android via OpenMedKit, React Native, plain CPUs, Apple Silicon, NVIDIA GPUs, the browser, and REST/gRPC services. No cloud. No vendor lock-in. No patient data leaving your network.

2,000+ models  ·  17 model-backed PII languages  ·  600+ PII checkpoints  ·  100% on-device  ·  Apache-2.0

English · 简体中文 · Español · Français · Deutsch · Italiano · Português · Nederlands · العربية · हिन्दी · తెలుగు · 日本語 · Türkçe · فارسی

---

See it in action

OpenMed runs entirely on the device; clinical text never leaves it. Here it is on iPhone, fully offline:

_{On iPhone via OpenMedKit: scan a clinical note, de-identify it, and extract clinical signals, all locally with Apple MLX. Nothing is uploaded.}

_{Real-time PII de-identification: the Nemotron Privacy Filter redacting names, addresses, IDs, and billing data from a clinical discharge packet, entirely on-device. (All values shown are synthetic.)}

---

30-second example

from openmed import analyze_text

result = analyze_text(
    "Patient started on imatinib for chronic myeloid leukemia.",
    model_name="disease_detection_superclinical",
)

for entity in result.entities:
    print(f"{entity.label:<12} {entity.text:<28} {entity.confidence:.2f}")
# DISEASE      chronic myeloid leukemia     0.98
# DRUG         imatinib                     0.95

A state-of-the-art clinical NER model running locally: no API key, no network call.

---

Why OpenMed?

	OpenMed	Cloud medical APIs
Runs on your device / servers	✅	❌
Patient data leaves your network	Never	Sent to the vendor
Cost	Free & open-source	Per-call pricing
Specialized medical models	2,000+	Limited
Model-backed PII languages	17	Varies
Offline / air-gapped	✅	❌
Apple Silicon (MLX) acceleration	✅	n/a
Native iOS / macOS apps	✅ via OpenMedKit	❌
Browser/WebGPU token classification	✅ via Transformers.js	Varies
Vendor lock-in	None (Apache-2.0)	Yes

• Specialized models: 2,000+ curated biomedical & clinical models, many outperforming proprietary stacks.
• HIPAA-aware de-identification: all 18 Safe Harbor identifiers, smart entity merging, format-preserving fakes.
• Runs everywhere: CPU, CUDA, Apple Silicon (MLX), iOS/macOS via OpenMedKit, Android/Kotlin, React Native, REST/gRPC services, and browser/WebGPU bundles via Transformers.js.
• One-line deployment: Python API, Dockerized REST service, or batch pipelines.
• Zero lock-in: Apache-2.0, your infrastructure, your data.

---

On-device on Apple: Swift, MLX & iOS

OpenMed is built to run where your data already lives. On Apple hardware it accelerates with MLX, and it ships straight into iPhone, iPad, and Mac apps through OpenMedKit: so PII detection and clinical extraction happen fully offline, on the device.

// Add OpenMedKit to your app
dependencies: [
    .package(url: "https://github.com/maziyarpanahi/openmed.git", from: "1.9.1"),
]

Expected result: Swift Package Manager resolves OpenMedKit and makes import OpenMedKit available to your app target.

• MLX runtime for PII token classification, the Privacy Filter family, experimental GLiNER-family zero-shot tasks, and Python MLX-LM text generation with Laneformer; includes a CoreML fallback path for supported token-classification artifacts.
• One model name, every platform: MLX model names automatically fall back to the matching PyTorch checkpoint on non-Apple hardware.
• Python on Apple Silicon too: pip install --upgrade "openmed[mlx]".

Guides: MLX backend · OpenMedKit (Swift) · CoreML export

_{MLX on Apple Silicon: 24–33× faster than CPU PyTorch for the Privacy Filter: median latency per inference step, lower is better.}

---

On-device on Android — Kotlin & ONNX Runtime Mobile

OpenMedKit also ships as a native Android/Kotlin library for local document intake, OCR handoff, PII redaction, and token-classification inference through ONNX Runtime Mobile. Mobile model repositories include stable tensor names, dynamic sequence axes, tokenizer files, labels, and Android-ready fp32, fp16, INT8, and optional .ort outputs.

Add the scoped JitPack repository in settings.gradle.kts:

dependencyResolutionManagement {
    repositories {
        google()
        mavenCentral()
        maven {
            url = uri("https://jitpack.io")
            content { includeGroup("com.github.maziyarpanahi") }
        }
    }
}

Then use the immutable OpenMed v1.9.1 release:

dependencies {
    implementation("com.github.maziyarpanahi:openmed:v1.9.1")
}

See the Android installation guide for local builds and publishing details.

val model = OpenMedKit.fromDirectory(modelDir)
val entities = model.analyzeText("Patient Alice Nguyen was seen in cardiology.")

• Android ONNX profile emits model.onnx, model_fp16.onnx, model_int8.onnx, tokenizer assets, labels, and openmed-onnx.json.
• ORT Mobile support records the minimal-build operator configuration when ONNX Runtime conversion tooling is installed.
• Kotlin parity tests keep tokenizer offsets, span boundaries, and decoder output aligned with the Python runtime.

Guides: Android ONNX export · Android span parity · OpenMedKit Android

The same ONNX model on Python CPU

from openmed import OnnxModel

model = OnnxModel.from_pretrained(
    "OpenMed/OpenMed-PII-ClinicalE5-Small-33M-v1-onnx-android"
)
entities = model("Patient Alice Nguyen was seen in cardiology.")

The same ONNX model in the browser

npm install openmed @huggingface/transformers

import { loadOnnxModel } from "openmed";

const model = await loadOnnxModel(
  "OpenMed/OpenMed-PII-ClinicalE5-Small-33M-v1-onnx-android",
);
const entities = await model("Patient Alice Nguyen was seen in cardiology.");

---

How it works

flowchart LR
    A["Clinical text"] --> B["OpenMed<br/>(100% on-device)"]
    B --> C["Medical entities"]
    B --> D["PII detected"]
    B --> E["De-identified text"]
    style B fill:#0D6E6E,stroke:#0A5656,stroke-width:2px,color:#ffffff
    style C fill:#D6EBEB,stroke:#0D6E6E,color:#0E1116
    style D fill:#F7DCD8,stroke:#C5453A,color:#0E1116
    style E fill:#F5E27A,stroke:#A9A088,color:#0E1116

Loading

Rendered result: a local clinical-text pipeline that returns medical entities, PII findings, and de-identified text without sending data to a cloud API.

---

Quick start

# Core + Hugging Face runtime (Linux, macOS, Windows; CPU or CUDA)
pip install --upgrade "openmed[hf]"

# Add the REST service
pip install --upgrade "openmed[hf,service]"

# Apple Silicon acceleration (MLX)
pip install --upgrade "openmed[mlx]"

Expected result:

Successfully installed openmed-...

Python API from openmed import analyze_text result = analyze_text( "Patient received 75mg " "clopidogrel for NSTEMI.", model_name= "pharma_detection_superclinical", ) print([(e.label, e.text) for e in result.entities]) Example output: [('DRUG', 'clopidogrel'), ('CONDITION', 'NSTEMI')]

REST service uvicorn openmed.service.app:app \ --host 0.0.0.0 --port 8080 Example output: INFO: Uvicorn running on http://0.0.0.0:8080 GET /health -> 200 OK GET /health POST /analyze POST /pii/extract POST /pii/deidentify

Batch from openmed import BatchProcessor p = BatchProcessor( model_name= "disease_detection_superclinical", group_entities=True, ) results = p.process_texts([...]) print(len(results), sum(len(r.entities) for r in results)) print([(e.label, e.text) for e in results[0].entities[:1]]) Example output: 3 7 [('DISEASE', 'leukemia')]

Browser / WebGPU

Package ONNX token-classification exports for in-browser inference through Transformers.js:

python -m openmed.onnx.convert \
  --model dslim/bert-base-NER \
  --output dist/example-onnx \
  --include-transformersjs

Example output:

Exported Transformers.js bundle to dist/example-onnx

import { pipeline } from "@huggingface/transformers";

const detector = await pipeline(
  "token-classification",
  "/models/openmed-pii/transformersjs",
  { device: "webgpu" },
);
const entities = await detector("Patient Casey Example called 212-555-0198.");
console.log(entities.slice(0, 2));

Example output:

[
  { entity: "NAME", word: "Casey Example", score: 0.99 },
  { entity: "PHONE", word: "212-555-0198", score: 0.98 },
]

Transformers.js export guide

Offline / air-gapped? Point model_name (or model_id) at a local directory and OpenMed loads it without contacting the Hugging Face Hub:

from openmed import OpenMedConfig, analyze_text

result = analyze_text(
    "Patient presents with chronic myeloid leukemia and Type 2 diabetes.",
    model_id="./models/OpenMed-NER-DiseaseDetect-SuperClinical-434M",
    config=OpenMedConfig(device="cpu"),
)
for entity in result.entities:
    print(f"{entity.label:<12} {entity.text:<28} {entity.confidence:.2f}")

Example output:

DISEASE      chronic myeloid leukemia     0.98
DISEASE      Type 2 diabetes              0.96

Because model_id points to a local directory, this example does not contact the Hugging Face Hub or any external model provider.

---

Models

A curated registry of specialized medical NER models; browse the full catalog.

Model	Specialization	Entity types	Size
disease_detection_superclinical	Disease & conditions	DISEASE, CONDITION, DIAGNOSIS	434M
pharma_detection_superclinical	Drugs & medications	DRUG, MEDICATION, TREATMENT	434M
pii_superclinical_large	PII & de-identification	NAME, DATE, SSN, PHONE, EMAIL, ADDRESS	434M
anatomy_detection_electramed	Anatomy & body parts	ANATOMY, ORGAN, BODY_PART	109M
gene_detection_genecorpus	Genes & proteins	GENE, PROTEIN	109M

---

Privacy: PII detection & de-identification

from openmed import extract_pii, deidentify

text = "Patient: John Doe, DOB: 01/15/1970, SSN: 123-45-6789"

# Extract PII with smart merging (prevents tokenization fragmentation)
result = extract_pii(text, model_name="pii_superclinical_large", use_smart_merging=True)
print([(e.label, e.text) for e in result.entities])

# De-identify with the method you need
print(deidentify(text, method="mask").deidentified_text)
print(deidentify(text, method="replace").deidentified_text)
print(deidentify(text, method="hash").deidentified_text)
print(deidentify(text, method="shift_dates", date_shift_days=180).deidentified_text)

Example output:

[('NAME', 'John Doe'), ('DATE', '01/15/1970'), ('SSN', '123-45-6789')]
Patient: [NAME], DOB: [DATE], SSN: [SSN]
Patient: Emily Chen, DOB: 03/22/1985, SSN: 456-78-9012
Patient: 6b8f...c4a1, DOB: 48b1...91de, SSN: 3f13...e912
Patient: John Doe, DOB: 07/14/1970, SSN: 123-45-6789

• Smart entity merging keeps 01/15/1970 whole instead of fragmenting it.
• Policy-aware pipelines add HIPAA/GDPR/research profiles, calibrated thresholds, signed audit reports, redaction previews, and minimum-necessary action selection.
• Faker-backed obfuscation with custom clinical-ID providers (CPF, CNPJ, BSN, NIR, Codice Fiscale, NIE, Aadhaar, Steuer-ID, NPI).
• HIPAA: all 18 Safe Harbor identifiers, configurable confidence thresholds.
• Batch and streaming PII: extract or de-identify across many documents with BatchProcessor(operation="extract_pii" | "deidentify", batch_size=16) or incremental streaming helpers.

_{Batch processing: up to 3.3× higher throughput on CPU and 2.2× on MLX vs. one document at a time.}

Complete PII notebook · Smart merging · Anonymization quickstart

Privacy Filter family: three model families on the OpenAI Privacy Filter architecture

Same model code (gpt-oss-style sparse-MoE transformer with local attention, sink tokens, RoPE+YaRN, tiktoken o200k_base), different training data. All route through the same extract_pii() / deidentify() API; only model_name= changes. openai/privacy-filter is a Hugging Face model identifier for local weights; using it here does not call the OpenAI API.

Variant	PyTorch (CPU + CUDA)	MLX (Apple Silicon)	MLX 8-bit
OpenAI Privacy Filter	openai/privacy-filter	OpenMed/privacy-filter-mlx	…-mlx-8bit
Nemotron-PII fine-tune	OpenMed/privacy-filter-nemotron	…-nemotron-mlx	…-nemotron-mlx-8bit
OpenMed Multilingual	OpenMed/privacy-filter-multilingual	…-multilingual-mlx	…-multilingual-mlx-8bit

from openmed import extract_pii

text = "Patient Sarah Connor (DOB: 03/15/1985) at MRN 4471882."

variants = {
    "baseline": extract_pii(text, model_name="openai/privacy-filter"),
    "nemotron": extract_pii(text, model_name="OpenMed/privacy-filter-nemotron"),
    "mlx": extract_pii(text, model_name="OpenMed/privacy-filter-mlx"),
}
print([(e.label, e.text) for e in variants["baseline"].entities])

Example output:

[('NAME', 'Sarah Connor'), ('DATE', '03/15/1985'), ('ID', '4471882')]

On non-Apple-Silicon hosts, MLX model names are automatically substituted with the matching PyTorch checkpoint (with a one-time warning): ship one model name, run anywhere. See Privacy Filter architecture & backend routing.

---

Multilingual PII (17 model-backed languages)

Extraction and de-identification support 17 supported PII language codes: ar, de, en, es, fr, he, hi, id, it, ja, ko, nl, pt, ro, te, th, and tr, with 600+ PII checkpoints in total. These are the model-backed PII language allow-list. OpenMed also includes validator-backed national-ID coverage for additional ID-only locales such as Polish, Latvian, Slovak, Malay, Filipino, and Danish.

See the per-language guide for each code's default PII model, Faker locale, and a before/after de-identification example.

python -c "from openmed import extract_pii; print([(e.label, e.text) for e in extract_pii('Dr. Pedro Almeida, CPF: 123.456.789-09, email: pedro@hospital.pt', lang='pt').entities])"

Example output:

[('NAME', 'Pedro Almeida'), ('ID', '123.456.789-09'), ('EMAIL', 'pedro@hospital.pt')]

Show per-language examples (Portuguese, Dutch, Hindi, Arabic, Japanese, Turkish)

from openmed import extract_pii

portuguese = extract_pii("Paciente: Pedro Almeida, CPF: 123.456.789-09, telefone: +351 912 345 678", lang="pt", use_smart_merging=True)
dutch      = extract_pii("Patiënt: Eva de Vries, BSN: 123456782, telefoon: +31 6 12345678", lang="nl", use_smart_merging=True)
hindi      = extract_pii("रोगी: अनीता शर्मा, फोन: +91 9876543210, पता: नई दिल्ली 110001", lang="hi", use_smart_merging=True)
arabic     = extract_pii("المريضة ليلى حسن، الهاتف +20 10 1234 5678، الرقم القومي 29801011234567.", lang="ar", use_smart_merging=True)
japanese   = extract_pii("患者 佐藤 花子、電話 +81 90 1234 5678、マイナンバー 1234 5678 9012.", lang="ja", use_smart_merging=True)
turkish    = extract_pii("Hasta Ayşe Yılmaz, telefon +90 532 123 45 67, TCKN 10000000146.", lang="tr", use_smart_merging=True)

for r in (portuguese, dutch, hindi, arabic, japanese, turkish):
    print([(e.label, e.text) for e in r.entities])

Example output:

[('NAME', 'Pedro Almeida'), ('ID', '123.456.789-09'), ('PHONE', '+351 912 345 678')]
[('NAME', 'Eva de Vries'), ('ID', '123456782'), ('PHONE', '+31 6 12345678')]
[('NAME', 'अनीता शर्मा'), ('PHONE', '+91 9876543210'), ('ADDRESS', 'नई दिल्ली 110001')]
[('NAME', 'ليلى حسن'), ('PHONE', '+20 10 1234 5678'), ('ID', '29801011234567')]
[('NAME', '佐藤 花子'), ('PHONE', '+81 90 1234 5678'), ('ID', '1234 5678 9012')]
[('NAME', 'Ayşe Yılmaz'), ('PHONE', '+90 532 123 45 67'), ('ID', '10000000146')]

---

REST API

A Docker-friendly FastAPI service with request validation, shared pipeline preload, and unified error envelopes.

pip install --upgrade "openmed[hf,service]"
uvicorn openmed.service.app:app --host 0.0.0.0 --port 8080

# or with Docker
docker build -t openmed:local .
docker run --rm -p 8080:8080 -e OPENMED_PROFILE=prod openmed:local

Example output:

INFO:     Uvicorn running on http://0.0.0.0:8080

curl -X POST http://127.0.0.1:8080/pii/extract \
  -H "Content-Type: application/json" \
  -d '{"text":"Paciente: Maria Garcia, DNI: 12345678Z","lang":"es"}'

Abbreviated example response:

{
  "text": "Paciente: Maria Garcia, DNI: 12345678Z",
  "entities": [
    {"text": "Maria Garcia", "label": "NAME", "confidence": 0.99, "start": 10, "end": 22},
    {"text": "12345678Z", "label": "ID", "confidence": 0.98, "start": 29, "end": 38}
  ],
  "model_name": "OpenMed/privacy-filter-multilingual"
}

Model lifecycle and service controls: free memory on demand with GET /models/loaded, POST /models/unload, and a keep_alive idle window; v1.8 also includes API-key/JWT auth, no-PHI request logging, tracing, gRPC, async jobs, webhooks, warm pools, dynamic batching, request coalescing, rate and concurrency limits, /livez, /readyz, and opt-in metrics:

OPENMED_SERVICE_KEEP_ALIVE=10m uvicorn openmed.service.app:app --host 0.0.0.0 --port 8080
curl -X POST http://127.0.0.1:8080/models/unload -H "Content-Type: application/json" -d '{"all":true}'

Example response:

{
  "unloaded": true,
  "released": {"models": 1, "tokenizers": 1, "pipelines": 1},
  "active_models": {}
}

See the full REST service guide.

---

Documentation

Full guides at openmed.life/docs.

Getting Started	Analyze Text	Model Registry
FAQ	Anonymization	Batch Processing
Configuration Profiles	REST Service	MLX Backend
Transformers.js Export	FHIR Interop	HL7 v2 De-identification
OpenMed 1.9.1 Release Notes	OpenMed 1.9.0 Release Notes	Examples
Release Streams	Generative Model Policy	Contributing
Security Policy	Compliance Posture

---

Meet the mascot

OpenMed's guardian is a fluffy Persian cat styled as a tiny Avicenna (Ibn Sina): the great Persian physician whose Canon of Medicine was the world's standard medical text for some 600 years. He keeps watch over the open book of medical knowledge, in a palette built around Persian turquoise (fīrūza): a local-first guardian for your most private data.

---

Contributing

Contributions welcome: bug reports, feature requests, and PRs alike. Please read the Contributing guide and our Code of Conduct first.

• Open an issue
• Contributing guide · Code of Conduct · Security policy
• Translations welcome: help complete the other-language READMEs linked in the switcher at the top.

---

Security

Found a vulnerability? OpenMed redacts PHI, so a redaction bypass or PHI/PII leak is a security issue: please report it privately, never as a public issue. See SECURITY.md for the responsible-disclosure policy and the private reporting form. Never include real patient data in a report.

---

Credits

OpenMed builds on excellent open-source work: particular thanks to OpenAI (the Privacy Filter architecture), NVIDIA (the Nemotron PII dataset), Hugging Face (transformers, Transformers.js & the model ecosystem), Apple (MLX), and the Faker maintainers.

License

Released under the Apache-2.0 License. Third-party asset notices are recorded in NOTICE.

Citation

@misc{panahi2025openmedneropensourcedomainadapted,
      title={OpenMed NER: Open-Source, Domain-Adapted State-of-the-Art Transformers for Biomedical NER Across 12 Public Datasets},
      author={Maziyar Panahi},
      year={2025},
      eprint={2508.01630},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2508.01630},
}

Expected result: BibTeX-compatible citation metadata for referencing OpenMed in papers, posters, and derived documentation.

---

Star History

If OpenMed is useful to you, a star helps others discover it.

---

Built by the OpenMed team

Website · Docs · X / Twitter · LinkedIn