app.py

from PIL import Image
import requests
import gradio as gr
from io import BytesIO

from transformers import BlipProcessor, BlipForConditionalGeneration

model_id = "Salesforce/blip-image-captioning-base"

model = BlipForConditionalGeneration.from_pretrained(model_id)
processor = BlipProcessor.from_pretrained(model_id)

def decode_base64_image(image_string):
  base64_image = base64.b64decode(image_string)
  buffer = BytesIO(base64_image)
  return  Image.open(buffer)

def launch(input_b64):
    image = decode_base64_image(input_b64)
    #inputsData = data.pop("inputs", data)
    # decode base64 image to PIL
    inputs = processor(image, return_tensors="pt")
    out = model.generate(**inputs)
    return processor.decode(out[0], skip_special_tokens=True)

iface = gr.Interface(launch, inputs=gr.inputs.Image(), outputs="text")
iface.launch()