app.py

from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline
import gradio as gr
import torch

nlp = pipeline('question-answering',
               model="databricks/dolly-v2-3b",
               torch_dtype=torch.bfloat16, 
               trust_remote_code=True, 
               device_map="auto")

def qnamodel(context,question):
  question_set = {'context':context,'question':question}
  results = nlp(question_set)
  return results['answer']

interface = gr.Interface(fn=qnamodel, 
                        inputs=[gr.inputs.Textbox(lines=7, label="Context"), gr.inputs.Textbox(lines=2, label="Question")],
                        outputs= gr.outputs.Textbox(label="Answer"),
                        title='Context Question Answering')
                        
interface.launch(inline=False)