app.py

import gradio as gr
import os
from huggingface_hub import InferenceClient
from textblob import TextBlob
import json
import time
import logging

# Set up logging
logging.basicConfig(level=logging.INFO)

# Get the API token from the environment variable
api_token = os.getenv('HUGGINGFACEHUB_API_TOKEN')

client = InferenceClient(
    model="Futuresony/future_ai_12_10_2024.gguf",
    token=api_token
)

def format_alpaca_prompt(user_input, system_prompt, history):
    """Formats input in Alpaca/LLaMA style"""
    history_str = "\n".join([f"### Instruction:\n{h[0]}\n### Response:\n{h[1]}" for h in history])
    prompt = f"""{system_prompt}
{history_str}

### Instruction:
{user_input}

### Response:
"""
    return prompt

def respond(message, history, system_message, max_tokens, temperature, top_p):
    formatted_prompt = format_alpaca_prompt(message, system_message, history)

    response = client.text_generation(
        formatted_prompt,
        max_new_tokens=max_tokens,
        temperature=temperature,
        top_p=top_p,
    )

    # ✅ Extract only the response
    cleaned_response = response.split("### Response:")[-1].strip()
    
    history.append((message, cleaned_response))  # ✅ Update history with the new message and response
    
    yield cleaned_response  # ✅ Output only the answer

demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
        gr.Slider(minimum=1, maximum=250, value=128, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.9, step=0.1, label="Temperature"),
        gr.Slider(minimum=0.1, maximum=1.0, value=0.99, step=0.01, label="Top-p (nucleus sampling)"),
    ],
)

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