OpenMed-NER-PharmaDetect-MultiMed-568M

What it does

This model detects the single entity type "Chemical or drug" in English clinical notes and biomedical literature. It reports an F1 score of 0.954 on the BC5CDR chemical benchmark, with 567M parameters and a maximum sequence length of 8194 tokens.

Best for

• Medication lists from clinical notes
• Chemical mentions in research papers
• Drug interaction screening in literature

What it detects (1 labels)

• Chemical or drug

Quality

Chemical Entity Recognition - Chemical entities from the BC5CDR dataset

Use it

pip install transformers torch

from transformers import pipeline
ner = pipeline("token-classification", model="OpenMed/OpenMed-NER-PharmaDetect-MultiMed-568M", aggregation_strategy="simple")
print(ner("Patient prescribed 10 mg aspirin for hypertension."))

pip install --upgrade "openmed[onnx-runtime]"

from openmed import OnnxModel
model = OnnxModel.from_pretrained("OpenMed/OpenMed-NER-PharmaDetect-MultiMed-568M-v1-onnx-android")
for entity in model("The patient started imatinib after the consultation."):
    print(entity.to_dict())

pip install "openmed[mlx]"

from openmed import analyze_text
from openmed.core.config import OpenMedConfig
result = analyze_text("Patient started imatinib in March.", model_name="OpenMed/OpenMed-NER-PharmaDetect-MultiMed-568M", config=OpenMedConfig(backend="mlx"))
for e in result.entities:
    print(e.label, e.text)