Awesome-AI4Med

We have renamed Medical_NLP to Awesome-AI4Med. If you need to view the previous content, you can visit backup!

Tip

Awesome-AI4Med aims to systematically curate research resources in the field of medical artificial intelligence (AI4Med), covering 🧠 Medical LLMs, 🩺 Medical MLLMs, 📚 datasets, and 📊 benchmarks. By automatically extracting trending models and datasets from a large corpus of research papers and open-source projects, this repository provides a clean, structured, and easy-to-navigate collection that helps researchers and developers 🚀 quickly track field progress, 🔍 locate relevant resources, and accelerate innovation in medical AI.

📣 You can view our repository on specialized disease LLMs at FreedomIntelligence/Awesome-Specialized-Medical-LLMs!

Table of Contents:

• 1. Medical LLMs
• 2. Medical MLLMs
• 3. Datasets
  • 3.1 Text
  • 3.2 Multimodal
• 4. Benchmarks

1. Medical LLMs

Model Name	Paper	Model Weight	Data	Code
HuatuoGPT	[arXiv]	[HF (7B)][HF (13B)]	[HF]	[Github]
Apollo	[arXiv]	[HF (7B)][HF (34B)][HF (72B)]	[HF]	[Github]
HuatuoGPT-II	[arXiv]	[HF (7B)][HF (13B)][HF (34B)]	[HF (PT)][HF (SFT)]	[Github]
HuatuoGPT-o1	[arXiv]	[HF (7B)][HF (8B)][HF (70B)]	[HF]	[Github]
MMedLM	[arXiv]	[HF (8B)][HF (70B)]	[HF]	[Github]
PMC-LLaMA	[arXiv]	[HF (13B)]	[HF]	[Github]
Med42-v2	[arXiv]	[HF (8B)][HF (70B)]	-	[Github]
OpenBioLLM	-	[HF (8B)][HF (70B)]	-	-
UltraMedical	[arXiv]	[HF (8B)][HF (70B)]	[HF]	[Github]
BioMistral	[arXiv]	[HF (7B)]	-	[Github]
MediTron	[arXiv]	[HF (7B)][HF (70B)]	[HF]	[Github]
Baichuan-M2	[arXiv]	[HF (32B)]	-	[Github]
Baichuan-M1	[arXiv]	[HF (14B)]	-	[Github]
MedAlpaca	[arXiv]	[HF (7B)][HF (13B)]	[Github]	[Github]
RadFM	[arXiv]	[HF (13B)]	[Github]	[Github]
MedReason	[arXiv]	[HF (8B)]	[HF]	[Github]
AlphaMed	[arXiv]	[HF (7B)][HF (8B)]	[HF]	-
Fleming-R1	[arXiv]	[HF (7B)][HF (32B)]	-	[Github]
MedGemma	[arXiv]	[HF (27B)]	-	[Github]

↥ back to top

2. Medical MLLMs

Model Name	Paper	Model Weight	Data	Code
HuatuoGPT-Vision	[arXiv]	[HF (7B)][HF 34B]	[HF]	[Github]
ShizhenGPT	[arXiv]	[HF (7B)][HF (32B)]	[HF (PT)][HF (SFT)]	[Github]
Lingshu	[arXiv]	[HF (7B)][HF 32B]	-	-
LLaVA-Med	[arXiv]	[HF (7B)]	[Github]	[Github]
BiMediX2	[arXiv]	[HF (8B)][HF (70B)]	[Github]	[Github]
Hulu-Med	[arXiv]	[HF (4B)][HF (7B)][HF (14B)][HF (32B)]		[Github]
HealthGPT	[arXiv]	[HF (3.8B)]	[HF]	[Github]
Med-Flamingo	[arXiv]	[HF (7B)]	-	[Github]
MedGemma	[arXiv]	[HF (4B)][HF (27B)]	-	[Github]
Med-R1	[arXiv]	[HF (2B)]	-	[Github]
BiomedGPT	[arXiv]	[HF (7B)]	[Github]	[Github]
MedVLM-R1	[arXiv]	[HF (2B)]	-	[Github]
UniMedVL	[arXiv]	[HF (14B)]	-	[Github]
Chiron-o1	[arXiv]	[HF (2B)][HF (8B)]	-	[Github]
MiniGPT-Med	[arXiv]	[HF]	[Github]	[Github]
Fleming-VL	[arXiv]	[HF (8B)][HF (38B)]	-	[Github]
VILA-M3	[arXiv]	[HF (3B)][HF (8B)][HF (13B)]	-	[Github]

↥ back to top

3. Datasets

3.1 Text

Data Name	Paper	Link	Quantity
Huatuo-26M	[arXiv]	[Github]	26,000,000
medical-o1-reasoning-SFT	[arXiv]	[HF]	25,000
ReasonMed	[arXiv]	[HF]	370,000

↥ back to top

3.2 Multimodal

Data Name	Paper	Link	Quantity
PubMedVision	[arXiv]	[HF]	1,294,062
Med-MAT	[arXiv]	[HF]	2,360,000
MedTrinity	[arXiv]	[HF]	25,000,000
OmniMedVQA	[arXiv]	[OpenXLab]	118,010
BioVFM	[arXiv]	[Github]	21,000,000
SLAKE	[arXiv]	[Website]	-
ChiMed-VL	[arXiv]	[HF]	580,014
BiomedParseData	[arXiv]	[HF]	6,800,000
SA-Med2D	[arXiv]	[HF]	20,000,000
IMIS	[arXiv]	[HF]	361,000,000
PreCT	[arXiv]	[HF]	160,000

↥ back to top

4. Benchmarks

Evaluation Framework	Code	Supported Benchmark
MedEvalKit	[Github]	（1）Multimodal：：MMMU-Medical-test, MMMU-Medical-val, PMC_VQA, OmniMedVQA, IU XRAY, MedXpertQA-Multimodal, CheXpert Plus, MIMIC-CXR, VQA-RAD, SLAKE, PATH-VQA, MedFrameQA；（2）Text-only：：MedQA-USMLE, MedMCQA, PubMedQA, Medbullets-op4, Medbullets-op5, MedXpertQA-Text, SuperGPQA, HealthBench, CMB, CMExam, CMMLU, MedQA-MCMLE；
MedBench	[Website]	Text-only：：Med-Exam, MedHC, MedMC, MedSpeQA, MedHG, MedDG, MedLitQA, IMCS-V2-MRG, DBMHG, CMB-Clin-extended, DDx-basic, DDx-advanced, MedTreat, CMeEE, CMeIE, CHIP-CDEE, CHIP-CDN, CHIP-CTC, SMDoc, MedSafety, DrugCA, CriID

↥ back to top