app.py

from transformers import pipeline
import gradio as gr

model = pipeline("text-generation")

def predict(prompt):
    completion = model(prompt)[0]["generated_text"]
    return completion
    
gr.Interface(fn=predict, inputs="text", outputs="text").launch()