import os
import torch
from flask import Flask, render_template, request, redirect, url_for
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
from peft import PeftModel

# --- Load Model & Tokenizer ---

base_model_name = "unsloth/llama-3.2-3b-bnb-4bit"
adapter_model_name = "aismaanly/ai_synthetic"

bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_use_double_quant=True,
    bnb_4bit_compute_dtype=torch.bfloat16
)

print("Loading base model...")
model = AutoModelForCausalLM.from_pretrained(
    base_model_name,
    quantization_config=bnb_config,
    device_map="auto"
)

print("Loading tokenizer...")
tokenizer = AutoTokenizer.from_pretrained(base_model_name)

print("Loading PEFT adapter...")
model = PeftModel.from_pretrained(model, adapter_model_name)
model = model.merge_and_unload()
print("Model ready!")

# --- Flask App ---

app = Flask(__name__)

@app.route("/", methods=["GET"])
def index():
    return render_template("index.html")

@app.route("/generate", methods=["POST"])
def generate():
    prompt = request.form["prompt"]

    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
    outputs = model.generate(**inputs, max_new_tokens=100)
    text = tokenizer.decode(outputs[0], skip_special_tokens=True)

    return redirect(url_for("result", generated_text=text))

@app.route("/result")
def result():
    generated_text = request.args.get("generated_text", "")
    return render_template("result.html", generated_text=generated_text)

if __name__ == "__main__":
    port = int(os.environ.get("PORT", 7860))
    app.run(host="0.0.0.0", port=port)