I’m trying to run AI Workbench remotely on a brev.dev VM. brev.dev lets me rent a GPU and I’d like to try some of my models on larger GPUs.
The steps are
- Spin up a
brev.devVM usually namedawesome-gpu-name - Download and run the brev CLI on my Mac
- Run the brev.dev login to generate credentials
- Run
brev shellto set up an SSH connection and the login credentials. Now I’m logged into my brev.dev VMawesome-gpu-name-hoston an ssh tunnel. - Exit the brev shell
- Try
ssh awesome-gpu-name-hostwhich sets up the fingerprints blah blah and does an ssh intoawesome-gpu-name-host - So now I have the ability to set up an SSH connection between my Mac and the
brev.devVM - Open AI Workbench
- Click on add a remote location
- I can enter the IP address associated with that
awesome-gpu-nameentry that I have in my .brev/ssh_config that I used to verify above?
What do I enter into the SSH Key File field. brev.dev uses a pem file. I can’t seem to figure out what file goes in this field.
Please tick the appropriate box to help us categorize your post
Bug or Error
Feature Request
Documentation Issue
Other