app_2.py

import gradio as gr
from huggingface_hub import InferenceClient


# client = InferenceClient("ruslanmv/Medical-Llama3-8B")
client = InferenceClient("microsoft/BioGPT-Large-PubMedQA")
# client = InferenceClient("mistralai/Mistral-7B-Instruct-v0.3") 

# cemilcelik/distilgpt2_pubmed
# microsoft/biogpt
# microsoft/BioGPT-Large
# BioGPT-Large-PubMedQA
# basic "mistralai/Mistral-7B-Instruct-v0.3"

def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    messages = [{"role": "system", "content": system_message}]

    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})

    messages.append({"role": "user", "content": message})

    response = ""

    for message in client.chat_completion(
        messages,
        max_tokens=max_tokens,
        stream=True,
        temperature=temperature,
        top_p=top_p,
    ):
        token = message.choices[0].delta.content

        response += token
        yield response


demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are a medical chatbot that helps doctors and pathologists with pathological issues. Be concise and answer the questions with given information.", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.1, step=0.1, label="Temperature"),
        gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.95,
            step=0.05,
            label="Top-p (nucleus sampling)",
        ),
    ],
)


if __name__ == "__main__":
    demo.launch()