Downloads · 30 days
17
9% of all-time downloads
ysn-ir/mistral-medical-chat-lora-v1
mistral-medical-chat-lora-v1 is a text generation model from ysn-ir. Use it when you need the model to write or continue text. It is set up for peft. The card lists the license as apache-2.0.
A medical question-answering LoRA trained to provide safe, cautious, general medical information, based on verified and well-structured Q&A data.
Downloads · 30 days
17
9% of all-time downloads
All-time downloads
180
Public
Repo size
254 MB
Likes
1
Public
Click a slice to open those files.
.safetensors336 MB · 78%
From the Hugging Face model README
A medical question-answering LoRA trained to provide safe, cautious, general medical information, based on verified and well-structured Q&A data.
This version focuses on explaining medical concepts, terminology, and common conditions, rather than conversation or patient-style dialogue (which appears in v2).
It responds to medical questions with:
mistralai/Mistral-7B-Instruct-v0.2This model does not perform diagnosis and does not provide medication dosing.
| Use Case | Allowed | Notes |
|---|---|---|
| Medical knowledge Q&A | ✅ | Education-focused |
| Explaining symptoms conceptually | ✅ | High-level guidance, no diagnosis |
| Anatomy, physiology, terminology explanations | ✅ | Good fit |
| Medical student study assistant | ✅ | Reliable structure, not clinical |
| Diagnosing medical conditions | ❌ | Must avoid guessing |
| Recommending specific treatments or doses | ❌ | High risk |
| Emergency care instructions | ❌ | Always refer to real professionals |
The model is trained to:
This aligns the model with safe educational medical use.
| Dataset | Size | Purpose |
|---|---|---|
keivalya/MedQuad-MedicalQnADataset | ~47k Q&A pairs | Medical education knowledge base |
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
base = "mistralai/Mistral-7B-Instruct-v0.2"
adapter = "ysn-ir/mistral-medical-chat-lora-v1"
tokenizer = AutoTokenizer.from_pretrained(base)
model = AutoModelForCausalLM.from_pretrained(base, device_map="auto")
model = PeftModel.from_pretrained(model, adapter)
prompt = "What causes iron deficiency anemia?"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
output = model.generate(**inputs, max_new_tokens=200)
print(tokenizer.decode(output[0], skip_special_tokens=True))
If the user reports:
→ The model must recommend immediate medical attention.
Hugging Face: https://huggingface.co/ysn-ir
Maintained by: ysn-ir
Released under Apache-2.0.
Permitted for:
Not permitted for clinical decision-making.
This v1 model:
v2 builds on this with improved patient-style conversation safety behavior.
Use responsibly.