app.py

import os
import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM

HF_TOKEN = os.environ.get("HF_TOKEN")
MODEL = "google/gemma-3-270m"

tokenizer = AutoTokenizer.from_pretrained(MODEL, token=HF_TOKEN)
model = AutoModelForCausalLM.from_pretrained(MODEL, token=HF_TOKEN)

def chat(message):
    inputs = tokenizer(message, return_tensors="pt")
    outputs = model.generate(**inputs, max_new_tokens=128)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return response

gr.Interface(
    fn=chat,
    inputs="text",
    outputs="text",
    title="Gemma 3 (270M)"
).launch()