You can deploy this model and start earning money today!

amank-root/
demo-ddi-1.5b-lora

Name: demo-ddi-1.5b-lora
Availability: InStock

drug-drug-interaction

medical

fine-tuned

Qwen/Qwen2.5-1.5B-Instruct

adapter

Views

No views yet

DDI Triage LoRA — amank-root/demo-ddi-1.5b-lora

LoRA adapter fine-tuned on the OpenEnv-DDI drug-drug interaction triage task.

Base model: Qwen/Qwen2.5-1.5B-Instruct
Task: Given a patient's medication list and labs, flag/monitor/ignore DDIs and suggest safer alternatives.
Training: 2 500 SFT rows (18 real + 480 synthetic cases), 2 epochs, r=8 LoRA.

Load

python

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-1.5B-Instruct", torch_dtype="auto", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("amank-root/demo-ddi-1.5b-lora")
model = PeftModel.from_pretrained(base, "amank-root/demo-ddi-1.5b-lora")
model = model.merge_and_unload()  # optional: merge for faster inference