How to get started with Pythagora
As highlighted on the wiki's homepage, Pythagora Pro is the recommended choice for most users. It's the quickest way to begin building your application, optimized to use the best LLM models available.
If you're a visual learner, this tutorial is also available in video form.
Install the Pythagora VS Code extension from the VS Code marketplace. You can also click on the extension icon in VS Code, then search for 'pythagora', then click the blue 'Install' button:
If you need help during installation, this quick YouTube video will guide you through the process. After installing the extension, click on the Pythagora icon on the left-hand extensions sidebar in VS Code:
Sign up for your Pythagora account using Google Authentication:
Access Pythagora's settings by clicking the gear icon:
On the settings page, you can check your current version of the Pythagora extension and GPT Pilot (the Pythagora Core version number). Here, you'll also see details such as whether Pythagora Pro credits are applied to your account, if an API key is configured, or if a notification indicates "You are out of credit" prompting setup actions.
You will be asked if you will like to develop with Pythagora's Local or Cloud development mode. This article explains more about both development options.
On the Pythagora settings page, the GPT Pilot path section displays the directory where Pythagora's Core files are installed.
If you selected to develop with Pythagora's Cloud mode, this section will show /pythagora/. If you chose Pythagora's Local development mode, you will need to click the Change button to select a new, empty directory on your machine.
To get started with Pythagora Pro, go to the Pythagora's settings page and click the Upgrade button. You'll receive an email with setup instructions. Once Pro is activated, restart VS Code and Pythagora should now display that you're on a Pro plan and how many remaining credits your account has.
Alternatively, if you prefer using your own API keys from a LLM provider, follow these instructions to use Pythagora with your own API key.
If you've signed up for Pythagora Pro, no further configuration or changes to the config.json file are needed. Pythagora Pro is already configured to use the most efficient LLMs from various providers, selecting the best model for each task.
Start working on your application! If you encounter issues, have questions, or want to provide feedback, join our Discord server.