Now using predictive text.

app.py

@@ -2,10 +2,14 @@
 # gradio demo                                                                                                     
 #                                                                                                                 
 import gradio as gr
+from transformers import pipeline
 
-def greet(name):
-    return "Hello " + name
+model = pipeline("text-generation")
 
-demo = gr.Interface(fn=greet, inputs="text", outputs="text")
+def predict(prompt):
+    completion = model(prompt)[0]["generated_text"]
+    return completion
 
-demo.launch()
+demo = gr.Interface(fn=predict, inputs="text", outputs="text")
+
+demo.launch()