# PubMed Pre-training

An open-source large language model for the medical domain based on the Mistral architecture, further pre-trained on PubMed Central texts

BioMistral is an open-source large language model optimized for the medical domain based on the Mistral architecture, further pre-trained on PubMed Central open-access text data, supporting multilingual medical question-answering tasks.

MedCPT is a model capable of generating biomedical text embeddings, particularly suitable for semantic search (dense retrieval) tasks.

BioELECTRA is a biomedical domain-specific language model pre-trained based on the ELECTRA framework, setting performance records on various biomedical NLP tasks

Named entity recognition model trained on NCBI Disease dataset and BC5CDR dataset, specialized in identifying disease entities in biomedical literature

A lightweight BERT model specifically designed for biomedical text mining, running 3 times faster than BERT-base while achieving comparable or superior performance to BioBERT/PubMedBERT