app.py

from flask import Flask, request, jsonify, render_template
from huggingface_hub import InferenceClient
import os

# Initialize the Flask app
app = Flask(__name__)

# Initialize the Hugging Face Inference Client
client = InferenceClient("Futuresony/future_ai_12_10_2024.gguf")

@app.route("/")
def home():
    # Render the HTML template
    return render_template("index.html")

@app.route("/message", methods=["POST"])
def fetch_message():
    data = request.json
    message = data.get("text", "")
    if not message:
        return jsonify({"error": "No input provided."}), 400

    # Model parameters
    parameters = {
        "temperature": 0.7,  # Adjust creativity
        "top_p": 0.9,       # Top-p sampling
        "max_new_tokens": 200,  # Limit response length
        "do_sample": True    # Enable sampling for varied responses
    }

    # Process the message using the Hugging Face model
    try:
        response = client.text_generation(
            message,
            parameters=parameters,
            timeout=300  # 5 minutes timeout
        )
        return jsonify({"response": response})
    except Exception as e:
        return jsonify({"error": str(e)}), 500

if __name__ == "__main__":
    # Use PORT environment variable or default to 7860
    port = int(os.getenv("PORT", 7860))
    app.run(host="0.0.0.0", port=port)