Quick Guide: Setting Up a Custom API in Cursor (Using y-api)

Showcase Built for Cursor
,
1

If you frequently run out of Cursor’s built-in Pro quota or simply prefer a pay-as-you-go workflow, hooking up a third-party OpenAI-compatible endpoint is usually the easiest solution.

I’ve been testing y-api (https://y-api.bestvirtualgoods.com) recently. A few practical highlights:

  • Free Models Available: Several models are completely free to use directly for everyday coding tasks and scripts.
  • 1:20 Top-up Ratio: For paid/advanced models, the current credit top-up ratio is 1:20, making token usage very cost-effective.
  • Standard OpenAI Protocol: Native compatibility—no plugins needed, just change the Base URL in Cursor.

Here is a straightforward 3-step setup guide:

Step 1: Get Your API Key

  1. Sign up/log in at https://y-api.bestvirtualgoods.com.
  2. Go to the dashboard/tokens section, create a new API Key, and copy it (sk-xxxx).

Step 2: Configure Base URL in Cursor

  1. Press Ctrl + Shift + J (Windows/Linux) or Cmd + Shift + J (macOS) to open Cursor Settings.
  2. Navigate to the Models tab on the left:

    • Toggle OpenAI API Key on, paste your key, and click Verify.

    • Under Override OpenAI Base URL, enter:

      https://api.y-api.bestvirtualgoods.com/v1

      (Make sure not to omit /v1 at the end).

Step 3: Add Your Models

  1. In the same Models section, click + Add Model.
  2. Enter the exact model name you want to use (you can check the supported list in the y-api dashboard, including both the free models and paid models).
  3. Toggle them on, and disable any unused models to prevent accidental calls.

Once configured, hit Ctrl/Cmd + K or Ctrl/Cmd + L in the editor to run a quick test prompt. If code streams back normally, you’re all set. You can use the free models for routine work and switch to the paid ones when dealing with complex tasks.