SVD Docker Image/Runpod.IO setup

[mnb3000]

Wanted to share my work on containerizing/templating SVD deployment for runpod.io instances (or your local machine, does not matter)

Here's a Dockerhub image, it has all of the dependencies required to run SVD and automatically downloads the weights to /workspace directory if it can not find them there, you need to mount a directory there.

latest or svd-base-xt tag auto-downloads both models, if you only want a specific model use svd-base/svd-xt tags respectively

https://hub.docker.com/repository/docker/mnb3000/runpod-stable-video-diffusion/ https://github.com/mnb3000/runpod-stable-video-diffusion

Detailed README with documentation/examples available on both Dockerhub/GitHub.

And here are automatic preconfigured templates for runpod instances:

• Base model: https://runpod.io/gsc?template=dove05wvcv
• XT model: https://runpod.io/gsc?template=0yuqqd2v24

Streamlit UI is available on port 3000 (configurable with $PORT env var) SVD-XT runs fine on RTX A6000 instances for $0.8/hr with discrete_t set to 8

[mnb3000]

Update 29.11:

• Dockerfile/shell source is now available in GitHub Repo
• Added $MODEL_MOUNTPOINT and $PORT env variables
• Removed nginx reverse proxying
• Added detailed instructions to run the image locally with directory mounting examples in README on Dockerhub/GitHub

Update 30.11:

• Replaced git clone with direct wget download for weights, as the repo was updated with diffusers weights, which are not used to run streamlit demo
• Added build instructions to GitHub README
• Fixed weights symlinks to checkpoints/ folder

The image is available on Dockerhub with detailed instructions for running it locally or with a cloud template