🩺 NurseGemma-2B-Merged (Ready to Run)

NurseGemma is a specialised nursing AI assistant fine-tuned on 7,500+ nursing care plans.
This is the merged version (LoRA Adapter + Base Model), making it ready for direct inference and API use.

It is designed to "think like a nurse", following the ADPIE nursing process.

🚀 How to Use

1. In Python (Transformers)

No need for peft or unsloth! It works like any standard model.

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "NurseCitizenDeveloper/NurseGemma-2B-Merged"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")

prompt = """<start_of_turn>user
You are a nurse. Create a care plan for a patient with pneumonia.<end_of_turn>
<start_of_turn>model
"""

inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0]))

2. Hugging Face Inference API (Free)

You can use this model directly with the free API (Serverless Inference).

API URL: https://router.huggingface.co/models/NurseCitizenDeveloper/NurseGemma-2B-Merged

🎯 Intended Use

• Education: Helping nursing students verify their care plans.
• Case Generation: Creating clinical vignettes for practice.
• Simulation: Roleplaying patient scenarios.

⚠️ Clinical Warning: This model is for educational and research purposes only. It is NOT a clinical decision support tool.

🧠 Training Data

Fine-tuned on NurseReason-7k (PubMedQA vignettes converted to Nursing Care Plans).

---

Built with ❤️ by nurses, for nurses.