import gradio as gr
from transformers import pipeline


model = pipeline(task="image-to-text", 
                model="Salesforce/blip-image-captioning-base")

def launch(input):
    output = model(input)
    return output[0]['generated_text']

def main():
    iface = gr.Interface(launch,
                    inputs=gr.Image(type='pil'),
                    outputs="text")
    
    iface.launch()

if __name__=="__main__":
    main()