static/server.py

from flask import Flask, request, jsonify
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

app = Flask(__name__)

model_id = "google/gemma-3-1b-it"  

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
    device_map="auto"
)

def generate(prompt):
    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)

    out = model.generate(
        **inputs,
        max_new_tokens=60,
        temperature=0.9,
        do_sample=True
    )

    return tokenizer.decode(out[0], skip_special_tokens=True)

@app.route("/narrate", methods=["POST"])
def narrate():
    data = request.json

    prompt = f"""
You are Myco, a mystical forest spirit narrator in a browser game.
Respond in 1 short poetic sentence (max 20 words).

Event: {data['event']}
Score: {data.get('score', 0)}
"""

    text = generate(prompt)
    return jsonify({"text": text})

app.run(port=5000)