app.py

import os
import requests
import gradio as gr

API_URL = "https://api-inference.huggingface.co/models/EleutherAI/gpt-neo-125M"
print("Loaded token:", os.environ.get('HUGGINGFACEHUB_API_TOKEN'))

HEADERS = {
    "Authorization": f"Bearer {os.environ['HUGGINGFACEHUB_API_TOKEN']}"
}

chat_history = []

def chat(user_message):
    chat_history.append({"role": "user", "content": user_message})
    payload = {
        "inputs": {
            "past_user_inputs": [m["content"] for m in chat_history if m["role"] == "user"][:-1],
            "generated_responses": [m["content"] for m in chat_history if m["role"] == "assistant"],
            "text": user_message
        },
        "options": {"use_cache": False}
    }

    response = requests.post(API_URL, headers=HEADERS, json=payload)
    response.raise_for_status()
    result = response.json()

    if isinstance(result, dict) and "generated_text" in result:
        reply = result["generated_text"]
    else:
        reply = result[0].get("generated_text", "")

    reply = reply.strip()
    chat_history.append({"role": "assistant", "content": reply})
    return reply

with gr.Blocks() as demo:
    chatbot_ui = gr.Chatbot()
    msg = gr.Textbox(
        placeholder="Type your message and press Enter",
        lines=1,
        show_label=False
    )
    msg.submit(fn=chat, inputs=msg, outputs=chatbot_ui)

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