app.py

import gradio as gr
from huggingface_hub import InferenceClient
import chardet

client = InferenceClient("HuggingFaceH4/zephyr-7b-beta", token="") #generate Access tokens

file_content = None

def respond(message, history, system_message, max_tokens, temperature, top_p, file=None):
    global file_content

    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 = ""

    if file:
        try:
            file_content = file.decode("utf-8")  
        except UnicodeDecodeError:
            result = chardet.detect(file)
            encoding = result['encoding']
            file_content = file.decode(encoding, errors='ignore')

    if "contenu du fichier" in message.lower() and file_content:
        response += f"Contenu du fichier :\n{file_content}"
        yield response
        return

    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

with gr.Blocks() as demo:
    gr.Image(value="logo-gaia.png", label="Logo")
    gr.ChatInterface(
        respond,
        additional_inputs=[
            gr.Textbox(value="You are a friendly Chatbot.", 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=0.7, step=0.1, label="Temperature"),
            gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
            gr.File(label="Télécharger un fichier", type="binary"),  
        ],
    )

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