# microsoft/BioGPT

Repository: https://github.com/microsoft/BioGPT
Canonical: https://ross.abutalabs.com/products/biogpt
Language: Python
License: MIT
License Family: permissive
Last push: 2024-07-25T10:33:53+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1479, "days_push": 769, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4488, forks 484 (observed 2026-08-28T04:08:51.290336+00:00)

## What it is
BioGPT is Microsoft's domain-specific generative Transformer language model pre-trained on biomedical text, with implementation code and pre-trained/fine-tuned checkpoints. It supports downstream tasks like biomedical question answering (PubMedQA) and relation extraction (BC5CDR).

## Use cases
- generate biomedical text with a pretrained transformer
- answer biomedical questions on PubMedQA
- extract drug-disease relations from biomedical literature
- fine-tune a GPT model on PubMed abstracts
- load BioGPT checkpoints from Hugging Face

## When to choose
- you need a domain-specific language model for biomedical NLP research
- you want to reproduce published BioGPT benchmark results on PubMedQA or relation extraction
- you need a research codebase built on fairseq for biomedical text generation

## When to avoid
- you need a production-ready LLM inference server or API
- you want a modern, actively maintained stack (it pins PyTorch 1.12 and fairseq 0.12)
- you need general-purpose chat or instruction-following capabilities

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, nlp, llm-training
- domain: bioinformatics, healthcare, large-language-models
- platform: python
- tags: biomedical-nlp, pretrained-models, transformer, text-generation, relation-extraction, question-answering, fairseq, natural-language-processing

## Member repositories
- microsoft/BioGPT (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:51.290336+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:20:29.627333+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/BioGPT (fetched 2026-08-28T04:08:51.290336+00:00, sha 620f97fba69a)
- Data as of 2026-08-30T08:39:29.467469+00:00.