Model Follower.

Up to date

Rebuilt
2026-10-25 06:54 UTC
Release radar
117 on watch
Hub snapshot
2026-10-25
Re-checked
daily · every number sourced and dated

Models › biogpt family

biogpt

Hugging Face ↗ Follow changes

microsoft MIT out 4 years ago

Text in, text out, 1K tokens of context.

Verified 2026-10-25 Rank #570 in LLMs · score 74.5 method v1 Sources 3 Downloads 30d 20.9k · #451 of 619 tracked

Can I use it?MITPermissivelicence file2026-10-24
Do people use it?20.9kdownloads, 30 days#451 of 619 we track · below the median+1 likes, downloads −31% this week315 likesHugging Face2026-10-24

Worth reading

Picked by rule from 3 tracked pieces: the lab's own post, the most shared, a field report, the latest.

BioGPT is integrated into the Hugging Face transformers library for direct pipeline usage.

From Microsoft bioGPT: pre-trained transformer for biomedical text · github.com · 2023-02-01
Latest · 4 years ago

Microsoft bioGPT: pre-trained transformer for biomedical text

BioGPT is a generative pre-trained transformer model for biomedical text generation and mining, implemented with PyTorch version 1.12.0 and Python version 3.10.

  • The pre-trained model checkpoint is available for download via a direct URL link.
  • BioGPT is integrated into the Hugging Face transformers library for direct pipeline usage.
github.com 2023-02-01 3 HN pts 762 words

3 best of 3 tracked

The 2 picks above, plus these 1. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.

  1. DiscussionMicrosoft BioGPT: biomedical LLM achieved human paritytwitter.com · 2023-02-06 · 2 HN pts

Where it's used

Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.

No API provider serves it, so its downloads are people running it themselves: 1 quantized builds and 100 Spaces.

  • As a service0no API provider listed
  • On own hardware1quantized builds
  • Built on10087 fine-tunes · 13 adapters · 0 merges
  • In apps100+Hugging Face Spaces

Top Spaces running it: kadirnar/Whisper_M2M100_BioGpt ♥ 43 · AnimaLab/bias-test-gpt-pairs ♥ 12 · epochs-demos/MedicalImagingApp ♥ 6 · VikramSingh178/MedicalImagingApplication ♥ 4

The story so far

Weights, every dated article, and repo commits on one rail.

3 dated articles and 19 repo commits so far

Teal is the weights (2022-11-20), amber triangles are 3 dated articles (each links to it), grey ticks are repo commits.

DecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOct2023-02-03: 3 repo commits2022-12-15: 1 repo commit2022-11-25: 4 repo commits2022-11-24: 1 repo commit2022-11-23: 3 repo commits2022-11-22: 1 repo commit2022-11-20: 6 repo commitsBioGPT: A language model pre-trained on large-scale biomedical literature · 2023-02-01Microsoft bioGPT: pre-trained transformer for biomedical text · 2023-02-01Microsoft BioGPT: biomedical LLM achieved human parity · 2023-02-06Weights 2022-11-20 · 2022-11-20Weights 2022-11-20now
DecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOct2023-02-03: 3 repo commits2022-12-15: 1 repo commit2022-11-25: 4 repo commits2022-11-24: 1 repo commit2022-11-23: 3 repo commits2022-11-22: 1 repo commit2022-11-20: 6 repo commitsBioGPT: A language model pre-trained on large-scale biomedical literature · 2023-02-01Microsoft bioGPT: pre-trained transformer for biomedical text · 2023-02-01Microsoft BioGPT: biomedical LLM achieved human parity · 2023-02-06Weights 2022-11-20 · 2022-11-20Weights 2022-11-20now

What people built on it

101 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.

101 derivatives on the Hub, most of them fine-tunes

From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.

1 quantized (Hugging Face model tree)87 fine-tunes (Hugging Face model tree)13 adapters (Hugging Face model tree)1 quantized87 fine-tunes13 adapters
1 quantized (Hugging Face model tree)87 fine-tunes (Hugging Face model tree)13 adapters (Hugging Face model tree)1 quantized87 fine-tunes13 adapters

All variants →

Reference

Everything else, folded: the full tables, every source.

What the card says

1 cited sentence

Summary generated, cited

BioGPT is a domain-specific generative Transformer language model pre-trained on large-scale biomedical literature.[1]

  1. “domain-specific generative Transformer language model pre…”

Generated 2026-10-07 15:12 UTC by parasail-gemma-4-26b-a4b-it; each numeral points at the fact it rests on. Not a verdict.

Facts

16 fields

Listed in LLMs

Facts

Pipeline tagtext-generation
Librarytransformers
ArchitecturesBioGptForCausalLM
Model typebiogpt
MoEno
Context length1,024 (config.json max_position_embeddings)
Languagesen
LicenseMIT (mit)
Hub snapshot2026-10-25

Servability

vLLM · SGLang

Servability

vLLM: not in registry · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: not in registry · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed

vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.

All variants

40 variants

Lineage

Original model Family: microsoft/biogpt 101 derivatives on the Hub

finetune: 87quantized: 1adapter: 13

Top variants (40 shown, ranked by downloads)

Relation
40 rows
Variant Relation Quant Params Band License Author Likes Downloads
asdasd123fasf/ajax12_finetuned finetune — — — mit asdasd123fasf 0 660
Sreehariiii/BioGPT-4bit-quantized-version quantized -4bit 356M S — Sreehariiii 0 566
baraqw/ajax12_finetuned finetune — — — mit baraqw 1 433
Medhatvv/biogpt_icd10_enhanced finetune — 425M S mit Medhatvv 0 426
joshlevitas/biogpt-finetuned-pathnotes finetune — — — mit joshlevitas 0 399
Sonatafyai/BioGPT_DocBot_SonatafyAI_V1 finetune — 347M S mit Sonatafyai 0 365
Ashish-wav/biogpt-ncbi-ner finetune — 347M S mit Ashish-wav 0 316
aprilyan/biogpt-finetuned-ner finetune — 347M S mit aprilyan 0 292
nmisra/biogpt-ncbi-disease-ner finetune — 347M S mit nmisra 0 243
Irisissocute/results finetune — 347M S mit Irisissocute 0 233
SweetZiyi/test-trainer finetune — 347M S mit SweetZiyi 0 221
vpadmak1/biogpt-ncbi-disease-ner finetune — 347M S mit vpadmak1 0 188
Raajameenakshi/biogpt-ncbi-ner finetune — 347M S mit Raajameenakshi 1 129
adishri8/biogpt-ncbi-disease-ner finetune — 347M S mit adishri8 0 172
nik548/bioGPT_finetuned_ncbi finetune — 347M S mit nik548 0 162
sabshr/biogpt-finetuned-ncbi-ner finetune — 347M S mit sabshr 0 153
Tomisin777/biogpt-finetuned-ner finetune — 347M S mit Tomisin777 0 136
pganeshbabu/biogpt-finetuned-ner finetune — 347M S mit pganeshbabu 0 126
ArcaDiana/my-biogpt-disease-ner finetune — 347M S mit ArcaDiana 0 122
kwoniris/gpt2gipgpt-finetuned-ner finetune — 347M S mit kwoniris 0 114
wolvram/biogpt-ner-ncbi finetune — 347M S mit wolvram 0 108
Raajameenakshi/meenu_biogpt-ncbi-ner finetune — 347M S mit Raajameenakshi 0 107
Sepideh2027/PathogenAgentAI-BioGPT-LoRA adapter — — — mit Sepideh2027 0 104
SaltySander/MOSAIC finetune — — — apache-2.0 SaltySander 2 0
isk31/biogpt_birads_classifier finetune — 347M S mit isk31 0 98
tequilajerry/Biogpt-NER finetune — 347M S mit tequilajerry 1 47
Kevin201217/bert-finetuned-ner finetune — 347M S mit Kevin201217 0 95
Xebec897/autotrain-2bj8e-oe91c finetune — 347M S other Xebec897 0 94
zmc0806/biogpt-finetuned-ncbi-ner finetune — 347M S mit zmc0806 0 85
maull04/biogpt_finetuning finetune — 347M S mit maull04 0 84
jise/biogpt-finetuned-ner finetune — 347M S mit jise 0 84
nadiamomtaz/biogpt-ncbi-ner finetune — 347M S mit nadiamomtaz 0 81
Wgaliga/Lab6_NER_Gary finetune — 347M S mit Wgaliga 0 80
rahularutla/biogpt-adr finetune — 347M S mit rahularutla 0 72
ars1m/medical_chatbot_model finetune — 347M S mit ars1m 0 72
Xebec897/autotrain-dndyw-o24f7 finetune — 347M S other Xebec897 0 70
seratonini/biogpt-ncbi-disease-ner finetune — 347M S mit seratonini 0 69
kwoniris/biogpt-ncbi-disease finetune — 347M S mit kwoniris 0 68
rahularutla/biogpt-ner finetune — 347M S mit rahularutla 0 66
MANAN-B94/biogpt-ner finetune — 347M S mit MANAN-B94 0 65

Repository changelog

19 commits

Repository changelog 19 commits

DateCommitSHA
2023-02-03 Update README.md eb0d815
2023-02-03 Update README.md ba48acd
2023-02-03 Fix architectures 8663af4
2022-12-15 Update README.md (#3) ef3fc6f
2022-11-25 updated citation cmt 73440ec
2022-11-25 updated model card examples 7be807d
2022-11-25 updated model card 2c5ef04
2022-11-25 model key changes 723c5ce
2022-11-24 Update config.json 65f02c7
2022-11-23 model d1412a1
2022-11-23 Create config.json ce3fe54
2022-11-23 Upload vocab.json d38b8d9
2022-11-22 Upload vocab.json f065e28
2022-11-20 Upload merges.txt 7a90c9e
2022-11-20 Upload 2 files 6e524bc
2022-11-20 Delete bpt_vocab.json 8032810
2022-11-20 Delete biogpt_merges.txt 1989dc2
2022-11-20 Upload 2 files 9bc24cb
2022-11-20 initial commit edc0707

Articles and coverage

3 pieces

Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency. Summaries are generated from a stored copy of the article text; every sentence cites a span of that text, and nothing is shown that failed the citation check.

Analysis2

BioGPT: A language model pre-trained on large-scale biomedical literature
marktechpost.com2023-02-01HN front pageHN 105 pts · 75 comments
Microsoft bioGPT: pre-trained transformer for biomedical text
github.com2023-02-01authorityHN 3 pts · 0 comments762 words
BioGPT is a generative pre-trained transformer model for biomedical text generation and mining, implemented with PyTorch version 1.12.0 and Python version 3.10. Pre-trained and fine-tuned checkpoints are available via URL or the Hugging Face Hub for tasks including question answering on PubMedQA and relation extraction on BC5CDR. The project is MIT-licensed and relies on fairseq version 0.12.0 for its implementation.written, reviewed, cited
  • The pre-trained model checkpoint is available for download via a direct URL link.
  • BioGPT is integrated into the Hugging Face transformers library for direct pipeline usage.
  • Contributors must sign a Contributor License Agreement to grant rights to their contributions.

Community1

Microsoft BioGPT: biomedical LLM achieved human parity
twitter.com2023-02-06HN 2 pts · 0 comments

Similar models

3 models

Similar models

  • opt-125m
    same task (text-generation)
  • gpt-j-6b
    same task (text-generation), permissive license
  • DialoGPT-medium
    same task (text-generation), permissive license

Sources

3 sources

Sources

Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:40 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:12 UTC · Last verified 2026-10-25

Follow biogpt

Watch this model

Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.