app.py

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

# Initialize the InferenceClient with your API key
x="hf_uYlR"
y="EMsDNbxQPJCSAHgwthrylHZZKKmGyg"
u=x+y
client = InferenceClient(api_key=f"{u}")

# Create a Flask app
app = Flask(__name__)

@app.route('/chat', methods=['POST'])
def chat():
    # Get user message from the request
    user_message = request.json.get('message')
    
    # Check if the user message is provided
    if not user_message:
        return jsonify({"error": "No message provided"}), 400
    
    # Create a single message list for the request
    messages = [{"role": "user", "content": user_message}]
    
    # Create the chat completion request with the current message
    response = client.chat.completions.create(
        model="Qwen/Qwen2.5-72B-Instruct", 
        messages=messages, 
        max_tokens=1024,
        stream=False  # Set stream to False to get the full response at once
    )
    
    # Get the assistant's response
    assistant_message = response.choices[0].message.content
    
    # Return the assistant's response as JSON
    return jsonify({"response": assistant_message})

if __name__ == '__main__':
    app.run(host="0.0.0.0", port=7860)