1.5 KiB
ModelScope text2video Extension for AUTOMATIC1111's StableDiffusion WebUI
WIP!!!
Auto1111 extension consisting of implementation of ModelScope text2video using only Auto1111 webui dependencies and downloadable models (so no logins required anywhere)
8gbs vram should be enough to run on GPU with low vram vae on at 256x256. (some opts are not working properly rn) But, 24 frames length 256x256 video definitely fits into 12gbs of NVIDIA GeForce RTX 2080 Ti. We will appreciate any help with this extension, especially pull-requests.
There is a known issue with ffmpeg stitching, if ffmpeg fails and it outputs something like 'tuple split failed', go to 'stable-diffusion-webui/outputs/img2img-images/text2video-modelscope' and grab the frames from there until it's fixed.
Prompt: flowers turning into lava
Test example:
Where to get the weights
All the parts of the model at HuggingFace (download your weights from there and put into stable-diffusion-webui/models/ModelScope/t2v):
https://huggingface.co/damo-vilab/modelscope-damo-text-to-video-synthesis/tree/main
Dev resources
HuggingFace space:
https://huggingface.co/spaces/damo-vilab/modelscope-text-to-video-synthesis
The model PyTorch implementation from ModelScope:
https://github.com/modelscope/modelscope/tree/master/modelscope/models/multi_modal/video_synthesis
Google Colab from the devs:
https://colab.research.google.com/drive/1uW1ZqswkQ9Z9bp5Nbo5z59cAn7I0hE6R?usp=sharing