app.py

import base64
import gradio as gr
import torch
import torchvision
from diffusers import DiffusionPipeline
import PIL.Image
import numpy as np
from io import BytesIO

ldm = DiffusionPipeline.from_pretrained("fusing/latent-diffusion-text2im-large")

generator = torch.manual_seed(42)
    
def greet(name):
    #prompt = "A squirrel eating a burger"
    prompt = name
    image = ldm([prompt], generator=generator, eta=0.3, guidance_scale=6.0, num_inference_steps=50)
    
    image_processed = image.cpu().permute(0, 2, 3, 1)
    image_processed = image_processed  * 255.
    image_processed = image_processed.numpy().astype(np.uint8)
    image_pil = PIL.Image.fromarray(image_processed[0])
    
    # save image as buffer
    buffered = BytesIO()
    image_pil.save(buffered, format="JPEG")
    img_str = base64.b64encode(buffered.getvalue())
    print(img_str.decode('utf-8'))
    return img_str.decode('utf-8')
    #return "Gello " + prompt + "!!"

image = gr.Image(type="pil", label="Your result")
iface = gr.Interface(fn=greet, inputs="text", outputs="text")
iface.launch()