api.py

from flask import Flask, request, jsonify
from app import generate_chat_completion
import time

app = Flask(__name__)

@app.route('/v1/chat/completions', methods=['POST'])
def chat_completions():
    data = request.json
    messages = data.get('messages', [])
    max_tokens = data.get('max_tokens', 560)
    temperature = data.get('temperature', 0.8)

    if not messages or not isinstance(messages, list):
        return jsonify({"error": "A valid 'messages' list is required."}), 400

    try:
        start_time = time.time()

        # Expecting plain role-content dicts (not Gradio tuples)
        result = generate_chat_completion(
            message_history=messages,
            max_tokens=max_tokens,
            temperature=temperature
        )

        # Get only the assistant's latest message
        assistant_msg = result[-1] if isinstance(result, list) else result
        elapsed = time.time() - start_time

        return jsonify({
            "model": "mistralai/Mistral-7B-Instruct-v0.2",
            "choices": [{
                "message": {
                    "role": "assistant",
                    "content": assistant_msg
                }
            }],
            "usage": {
                "generation_time": round(elapsed, 2)
            }
        })

    except Exception as e:
        return jsonify({"error": str(e)}), 500

@app.route('/')
def health_check():
    return "LLM API is running", 200

if __name__ == '__main__':
    app.run(host='0.0.0.0', port=8081)