Downloads · 30 days
5
7% of all-time downloads
medgemma-medical-qa-lora
samandar1105/medgemma-medical-qa-lora
medgemma-medical-qa-lora is a question answering model from samandar1105. Use it when the input is a question plus a passage. It is set up for peft. The card lists the license as other.
Fine-tuned with QLoRA on the MedQuAD dataset for generative medical question answering. Intended to be used as the generator in a RAG pipeline — pair with a retriever (e.g. embeddings over epfl-llm/guidelines) for gro…
All-time downloads
68
Public
Repo size
296 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors131 MB · 80%
At a glance
- Task
- Question Answering
- Library
- peft
- License
- other
- Access
- Public
- Created
- Jul 6, 2026
- Updated
- Jul 6, 2026
- SHA
- 6499bf66
Try a prompt
Base models
Datasets
Model card
From the Hugging Face model README
MedGemma-4B fine-tuned for Medical Q&A (LoRA adapter)
Fine-tuned with QLoRA on the MedQuAD dataset for generative medical question answering.
Intended to be used as the generator in a RAG pipeline — pair with a retriever
(e.g. embeddings over epfl-llm/guidelines) for grounded, up-to-date answers.
⚠️ Disclaimer
This model is a research/portfolio project. It is not a certified medical device and must not be used for real clinical decision-making without human medical professional oversight.
Usage
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer
base = AutoModelForCausalLM.from_pretrained("google/medgemma-4b-it", device_map="auto")
model = PeftModel.from_pretrained(base, "samandar1105/medgemma-medical-qa-lora")
tokenizer = AutoTokenizer.from_pretrained("samandar1105/medgemma-medical-qa-lora")
- Task
- Question Answering
- Library
- peft
- License
- other
- Languages
- en
- Created
- Jul 6, 2026
- Updated
- Jul 6, 2026