app.py

import gradio as gr
from transformers import AutoTokenizer, TFAutoModelForCausalLM

def get_response(text):
    tokenizer = AutoTokenizer.from_pretrained("gpt2")
    model = TFAutoModelForCausalLM.from_pretrained("gpt2")
    model.config.pad_token_id = model.config.eos_token_id
    inputs = tokenizer([text], return_tensors="tf")

    generated = model.generate(**inputs, do_sample=True, seed=(42, 0),  max_new_tokens=20, temperature=.1)
    return tokenizer.decode(generated[0])

iface = gr.Interface(fn=get_response, inputs="text", outputs="text")
iface.launch()