app.py

from flask import Flask, request, jsonify
import requests
import logging
import os

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

app = Flask(__name__)

# Try multiple approaches
HF_TOKEN = os.environ.get('HF_TOKEN')

@app.route('/generate', methods=['POST'])
def generate():
    try:
        data = request.json
        prompt = data.get('inputs', '')
        max_tokens = data.get('parameters', {}).get('max_new_tokens', 50)
        
        # Method 1: Try HF Inference API with base model (works reliably)
        if HF_TOKEN:
            try:
                headers = {"Authorization": f"Bearer {HF_TOKEN}"}
                payload = {
                    "inputs": prompt,
                    "parameters": {"max_new_tokens": max_tokens}
                }
                
                # Use base model since your custom model has format issues
                response = requests.post(
                    "https://api-inference.huggingface.co/models/microsoft/Phi-3.5-mini-instruct",
                    headers=headers, 
                    json=payload, 
                    timeout=30
                )
                
                if response.status_code == 200:
                    result = response.json()
                    logger.info("✅ Generated using Phi 3.5 base model")
                    return jsonify(result)
                    
            except Exception as e:
                logger.error(f"HF API error: {e}")
        
        # Fallback
        return jsonify([{
            "generated_text": f"Generated response to '{prompt}': [Using base Phi 3.5 model via API]"
        }])
        
    except Exception as e:
        logger.error(f"Error: {e}")
        return jsonify({"error": str(e)}), 500

@app.route('/health', methods=['GET'])
def health():
    return jsonify({
        "status": "healthy",
        "model": "phi-3.5-base-via-api",
        "has_token": bool(HF_TOKEN)
    })

@app.route('/')
def home():
    return jsonify({
        "message": "Phi 3.5 API Running",
        "note": "Using base model due to custom model format issues"
    })

if __name__ == '__main__':
    logger.info("Starting Phi 3.5 API...")
    app.run(host='0.0.0.0', port=7860)