app.py

import gradio as gr
from transformers import pipeline

# Rename 'text' to 'image'
def do_action(image):
    # Use the blip model
    pipe = pipeline("image-to-text", model="Salesforce/blip-image-captioning-large")

    result = pipe(image)

    return result[0]['generated_text']

# input image, output text
# can't use inputs="image"
# instead use inputs=gr.Image(type='pil')
iface = gr.Interface(fn=do_action, inputs=gr.Image(type='pil'), outputs="text")
iface.launch()