BC5CDR chemical entity focus
The model card describes chemical entity recognition over 1,500 PubMed abstracts with 4,409 annotated chemical entities, emitting B-CHEM and I-CHEM labels for drugs, compounds, and therapeutic substances.
Open Source Model Profile · OpenMed
OpenMed-NER-PharmaDetect-SuperClinical-434M
OpenMed-NER-PharmaDetect-SuperClinical-434M is a 434M-parameter DeBERTa token-classification model from OpenMed. Its card documents BC5CDR chemical entity recognition.
- Publisher
- OpenMed
- Task
- token-classification
- Model type
- deberta-v2
- License
- apache-2.0
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
OpenMed-NER-PharmaDetect-SuperClinical-434M is published by OpenMed as a DeBERTa-family token-classification model. The captured configuration identifies DebertaV2ForTokenClassification and Safetensors metadata reports 434,015,235 parameters, or about 434M. The model card describes a BC5CDR_CHEM chemical-entity recognizer emitting B-CHEM and I-CHEM labels.
Recorded capabilities
Publisher-reported 0.9614 F1
According to the model card, the model reports F1 0.9614, precision 0.9520, recall 0.9710, and accuracy 0.9892 on BC5CDR_CHEM.
Transformers pipeline and batching docs
The model card documents Hugging Face pipeline use with aggregation_strategy simple, plus batch_size guidance for CPU, single-GPU, and high-end GPU processing.
Use cases in the source record
- Clinical text mining tasks named in the card, including drug-interaction detection, medication extraction, adverse-event monitoring, literature mining, and biomedical knowledge-graph construction.
- Batched biomedical NER pipelines over clinical notes and research text using the documented Transformers pipeline and batch-size workflow.
Limitations and unknowns
- Performance and ranking figures are publisher-reported on BC5CDR_CHEM and are not Ethen-measured results.
- Provider state is historical snapshot data and should be refreshed before being presented as current.
- No context-window, VRAM, quantization, or pricing figures were extracted.
- Captured architecture is DebertaV2ForTokenClassification while the card names deberta-v3-large; both labels were preserved without inferring equivalence.
Source and provenance
Source: OpenMed/OpenMed-NER-PharmaDetect-SuperClinical-434M
Captured: Unknown. Processed: 2026-09-07T19:34:35.234805+00:00.
🧬 OpenMed-NER-PharmaDetect-SuperClinical-434M Specialized model for Chemical Entity Recognition - Chemical entities from the BC5CDR dataset 📋 Model Overview This model is a state-of-the-art fine-tuned transformer engineered to deliver enterprise-grade accuracy for chemical entity recognition - chemical entities from the bc5cdr dataset. This specialized model excels at identifying and extracting biomedical entities from clinical texts, research papers, and healthcare documents, enabling applications such as drug interaction detection , medication extraction from patient records , adverse event monitoring , literature mining for dru…
F001F002F003F004F005F006F007F009F010F011F012F013F014F015F020F022F023