Using a remote system to serve AI Code helper with LM Studio & VS Code Continue extension (LLM)

LM Studio remote coding assistant.md

I'm using a Windows machine with an RTX 3070 for this.

on the windows side

1. install LM Studio
2. install and start OpenSSH server
3. download a model - ideally a code related one
4. from the server tab a) select the model at top b) accept the updated prompt if needed c) click Start Server

On the development machine

First from your development machine SSH into the other machine and port-forward the LM Studio port

ssh -L 1234:127.0.0.1:1234 desktop-e3rmgvg\\[email protected]

be sure to adjust machine name, user name and IP get machine name and user name from a non-admin powershell prompt with whoami get the IP from ipconfig

Add something like the following to the continue config.

    {
      "title": "LM STUDIO3",
      "provider": "openai",
      "model": "deepseek-ai_deepseek-coder-6.7b-base",
      "apiBase": "http://localhost:1234/v1/"
    },

Install the continue extension in VS Code

Read the docs for usage https://continue.dev/docs/how-to-use-continue#introduction in short though:

1. highlight code
2. cmd+shift+m
3. ask a question
4. to have it edit code prepend your question with /edit