app.py

import os
import pathlib

import gradio as gr
from huggingface_hub import HfApi, snapshot_download
from modelscope.outputs import OutputKeys
from modelscope.pipelines import pipeline

api = HfApi()
model_dir = pathlib.Path("weights")
snapshot_download(
    "damo-vilab/modelscope-damo-text-to-video-synthesis",
    repo_type="model",
    local_dir=model_dir,
    use_auth_token=os.environ["HUB_TOKEN"],
)

pipe = pipeline("text-to-video-synthesis", model_dir.as_posix())


def generate_video(prompt):
    text = {"text": prompt}
    output = pipe(text)[OutputKeys.OUTPUT_VIDEO]

    filename = "output.mp4"
    cwd = os.getcwd()
    video_path = os.path.join(cwd, filename)

    with open(video_path, "wb") as video_file:
        video_file.write(output)

    return video_path


iface = gr.Interface(fn=generate_video, inputs="text", outputs="video")

iface.launch()