Prediction Guard powered Code Assistant

Continue.dev is an open-source AI code assistant available as an extension for both VS Code and JetBrains. It is flexible, supporting various model providers. The best part? You can quickly and easily set up Prediction Guard as a provider by following these simple steps:

Step 1: Install the Continue.dev extension

You can install the Continue.dev extension from the VS Code Marketplace or from the JetBrains Plugins Marketplace.

Step 2: Add a provider

Open the Continue.dev chat interface by clicking the Continue.dev icon on the left panel in VS Code. In the chat tab, locate the dropdown that lists all the available providers. Click on + Add Chat Model. This will open a new tab to configure your custom provider.

Step 3: Modify config.json

In the newly opened tab, click on the config file hyperlink to open the file in VS Code. Add the following configuration to the file::

{
      "title": "Prediction Guard",
      "provider": "openai",
      "model": "<Model>",
      "systemMessage": "You are an expert software developer. You give helpful and concise responses.",
      "apiKey": "<Your PG API key>",
      "apiBase": "<Your PG API URL>"
    }

A list of models supported by Prediction Guard can be found in the documentation. Replace Model with the model you want to use and Your API key with your API key. Save the file.

For tabautocomplete modify the config.json like this:

"tabAutocompleteModel": {
    "title": "Prediction Guard",
    "provider": "openai",
    "model": "",
    "apiKey": "<Your PG API Key>",
    "apiBase": "<Your PG API URL>",
    "useLegacyCompletionsEndpoint": false
  }

Ensure that the Autocomplete is enabled in the settings.

Start Coding

Once saved, Prediction Guard will be available as a provider in Continue.dev, and you can start using it right away.

Happy Coding!