CloveAI/clov-medchat

Use this model

!pip install -q --upgrade bitsandbytes transformers accelerate

from transformers import pipeline

pipe = pipeline("text-generation", model="alanjoshua2005/alan-mistral-finetuned")

user_input = input("Enter your medical question or prompt: ")

prompt = (
    f"""Imagine you are a helpful medical chatbot. Respond based on the user input below:

<s>[INST] {user_input} [/INST]

Please provide your answer in **structured Markdown format**. Follow these rules:
- Complete the answer fully; do not stop mid-sentence
- Use emojis to highlight key points
- Use horizontal lines (---) to separate sections
- Use bullet points and numbered lists where appropriate
- Use tables if necessary to organize information clearly
- Explain medical terms in simple words
- Do NOT include any links, URLs, or image references
- Make the response easy-to-read and informative
"""
)

result = pipe(
    prompt,
    max_new_tokens=512,       
    do_sample=True,
    temperature=0.7,
    top_p=0.9,
    repetition_penalty=1.1    
)

generated_text = result[0]["generated_text"]
response = generated_text.replace(prompt, "").strip()

print(response)

---

Model Details

• Developed by: Alan Joshua
• Model type: Text-Generation
• Language(s): English
• License: MIT
• Finetuned from model: unsloth/mistral-7b-instruct-v0.2-bnb-4bit
• Dataset: ruslanmv/ai-medical-chatbot

---

Model Description

This model is a medical chatbot fine-tuned on the ruslanmv/ai-medical-chatbot dataset using LoRA adapters on the Mistral 7B instruct model (4-bit). It is designed to provide accurate, easy-to-understand medical information in English.

Key features of this model include:

• Structured Markdown responses: Answers are formatted using bullets, numbered lists, tables, and horizontal lines for readability.
• Clear explanations: Medical terms are explained in simple words for users of all backgrounds.
• Emojis: Used to highlight key points and make responses more engaging.
• No links or images: Ensures responses remain text-only for safe, direct answers.
• Complete answers: Designed to generate full, coherent responses without cutting off mid-sentence.

This model is suitable for educational purposes, healthcare awareness, and interactive Q&A applications. It is not a substitute for professional medical advice. Always verify information with a qualified healthcare provider.