[Solved] Sd-webui-text2video doesn't generate any videos

I successfully installed stable-diffusion-webui-1.7.0 along with its text to video extension from sd-webui-text2video. Also, I can generate image from text as well without a problem.

However, when attempting to generate a video through text, the ui gets stuck after clicking the Generate button, which switches to interrupt and Skip buttons, then no response. The terminal also do not have any progress bar or error messages.

The log messages in terminal are as below

...
Startup time: 5.1s (prepare environment: 0.2s, import torch: 1.7s, import gradio: 0.4s, setup paths: 1.2s, other imports: 0.3s, load scripts: 0.5s, create ui: 0.3s, gradio launch: 0.4s).
Applying attention optimization: InvokeAI... done.
Model loaded in 2.6s (load weights from disk: 0.4s, create model: 0.4s, apply weights to model: 0.9s, apply float(): 0.7s, calculate empty prompt: 0.1s). 
// no progress bar. no more message

In models/ModelScope/t2v/, there exists files

configuration.json  open_clip_pytorch_model.bin  text2video_pytorch_model.pth  VQGAN_autoencoder.pth

And models/VideoCrafter/ contains model.ckpt file.

My python3 version 3.10.14. Launch the webui.sh with export COMMANDLINE_ARGS="--skip-torch-cuda-test --no-half" params.

How should I fix this? Thanks

Ok I just found a workaround. Here it describes how to fix the problem.

  1. Edit extensions/sd-webui-text2video/javascript/t2v_progressbar.js
  2. Paste the following function below another function.
function setSubmitButtonsVisibility(tabname, showInterrupt, showSkip, showInterrupting) {
    gradioApp().getElementById(tabname + '_interrupt').style.display = showInterrupt ? "block" : "none";
    gradioApp().getElementById(tabname + '_skip').style.display = showSkip ? "block" : "none";
}

This topic was automatically closed 12 hours after the last reply. New replies are no longer allowed.