OpenMed-ZeroShot-NER-Oncology-Multi-209M

What it does

This model performs zero-shot named-entity recognition on English biomedical text, identifying 16 entity types such as Cancer, Gene or gene product, and Cell. It is built on the GLiNER architecture with 209M parameters and reports an F1 score of 0.75 on the BioNLP 2013 Cancer Genetics test set.

Best for

• Cancer pathway curation from literature
• Driver gene analysis in tumor biology
• Precision oncology literature mining

What it detects (16 labels)

Trained on the label below; you can pass any other label names at inference time.

• Amino acid
• Anatomical system
• Cancer
• Cell
• Cellular component
• Developing anatomical structure
• Gene or gene product
• Immaterial anatomical entity
• Multi-tissue structure
• Organ
• Organism
• Organism subdivision
• Organism substance
• Pathological formation
• Simple chemical
• Tissue

Quality

Cancer Genetics - Cancer-related genetic entities

Use it

pip install gliner

from gliner import GLiNER
model = GLiNER.from_pretrained("OpenMed/OpenMed-ZeroShot-NER-Oncology-Multi-209M")
print(model.predict_entities("Patient with type 2 diabetes and hypertension.", ["disease"]))

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-ZeroShot-NER-Oncology-Multi-209M", config=OpenMedConfig(backend="mlx"))
for e in result.entities:
    print(e.label, e.text)