Use Invoke AI - A creative engine for Stable Diffusion on Jarvislabs out of the box in less than 45 seconds ⏰. It comes with 15 preloaded models.
Get started in 2 simple steps.
- Launch an instance
- Start with your creative journey
Launch an instance
Start a instance with framework as Invoke AI. Choose required GPUs and storage.
Access the web UI
Once the instance is up and running, right click on your running instance and select the API endpoint.
This may take a few seconds to load. You can start with your creative journey.
Launch/Modify the parameters used to launch
When the instance is started, we start the
/home/invokeai/launch_invoke.sh from inside a tmux session. You can check that from the JupyterLab terminal.
From the terminal run the below command
tmux attach -t invoke
The output would look somthing like this.
If you do a 'CTRL + C', the application will stop. You can launch by running the script.
The script contains the below steps
invokeai --web --host 0.0.0.0 --port 6006 --cors $DNS
You can learn more on how to use the UI from the official page.
Find your favourite open source model missing, let us know on chat option or drop us an email at email@example.com. We will be happy to add it.
We are not the creators of invokeai 😊, we are just making it easier to use with Jarvislabs GPU instances. If you are stuck somewhere, we will try our best to help. You can always ping us using the chat.