app.py

import gradio as gr
from diffusers import DiffusionPipeline
from transformers import AutoTokenizer

# Load the tokenizer
tokenizer = AutoTokenizer.from_pretrained("./")

# Load the pre-trained model
pipe = DiffusionPipeline.from_pretrained("./", tokenizer=tokenizer)

# Function to generate video from text prompt
def generate_video(prompt):
    # Generate image from prompt
    image = pipe(prompt).images[0]
    # Generate video from image
    video = pipe(image=image).videos[0]
    return video

# Define Gradio interface
iface = gr.Interface(
    fn=generate_video,
    inputs=gr.Textbox(label="Enter your prompt"),
    outputs=gr.Video(label="Generated Video"),
    live=True,
)

# Launch the interface
iface.launch()