Cursor IDE

Cursor is an AI-powered code editor built on VS Code. FloTorch integrates seamlessly with Cursor IDE using its native support for OpenAI-compatible APIs. This allows you to connect Cursor directly to the FloTorch Gateway by configuring the gateway URL, API key, and your custom model name.

Native FloTorch Configuration in Cursor

Cursor provides built-in support for custom AI models through its Settings interface. You can configure FloTorch directly without installing any additional extensions.

Setup Steps

1. Open Cursor Settings

   • Navigate to Models

2. Configure OpenAI settings:

   • Enable OpenAI API Key and enter your FloTorch API key (Create API Key)
   • Enable Override OpenAI Base URL and set it to: https://gateway.flotorch.cloud/openai/v1

3. Add a new model:

   • In the model search box, type your FloTorch custom model name (Create Model)
   • The configuration is saved automatically

4. Select the FloTorch Model

   • Open Cursor’s AI Chat panel (Ctrl+L / Cmd+L)
   • Click the model selector dropdown
   • Select your configured FloTorch model

Alternative: Using OpenAI-Compatible Extensions

Since Cursor is built on VS Code, it supports VS Code extensions. Any extension that supports OpenAI-compatible APIs can be configured to work with FloTorch.

1. Roo Code

Extension: Roo Code

Setup Steps

1. Open Extensions in Cursor and install Roo Code, or install directly from the marketplace.

2. Open Roo Code (sidebar or panel).
3. Create a Roo Code account and redirect to Cursor.
4. Navigate to Settings → Providers.
5. Under API Provider, select OpenAI Compatible.
6. Configure:
   • Base URL: https://gateway.flotorch.cloud/openai/v1
   • API Key: Your FloTorch API key (Create API Key)
   • Model ID: Your FloTorch custom model name (Create Model)
7. Save the settings and select the model for chat and code actions.

2. Continue

Extension: Continue

Setup Steps

1. Install Continue from the Extensions Marketplace, or install directly from the marketplace.

2. Open Continue (from the sidebar or Command Palette).
3. Navigate to Settings → Configs.
4. Open Local Config (this opens configs.yaml).
5. Add the FloTorch model configuration:

name: Local Config
version: 1.0.0
schema: v1
models:
  - name: FloTorch Gateway
    provider: openai
    model: YOUR_FLOTORCH_MODEL_NAME
    apiBase: https://gateway.flotorch.cloud/openai/v1
    apiKey: YOUR_FLOTORCH_API_KEY
    roles:
      - chat
      - edit
      - apply

6. Replace YOUR_FLOTORCH_MODEL_NAME (Create Model) and YOUR_FLOTORCH_API_KEY (Create API Key) with your actual values, save, and select this model in Continue.

3. Kilo

Extension: Kilo

Setup Steps

1. Open Extensions in Cursor (Ctrl+Shift+X).
2. Search for Kilo in the Extensions Marketplace and install it, or install directly from the marketplace.

3. Open Kilo (from the sidebar or Command Palette).
4. Navigate to Settings → Providers.
5. Under API Provider, select OpenAI Compatible.
6. Configure:
   • Base URL / Endpoint: https://gateway.flotorch.cloud/openai/v1
   • API Key: Your FloTorch API key (Create API Key)
   • Model: Your FloTorch custom model name (Create Model)
7. Save and start using the model in Kilo.

---

Summary

FloTorch integrates with Cursor IDE both natively and through OpenAI-compatible extensions. Use the native configuration for the simplest setup, or install any of the extensions above (Roo Code, Continue, Kilo) with the same FloTorch gateway URL, API key, and model name. Other OpenAI-compatible extensions—such as Cline, vscode-openai, and similar tools that let you set a custom base URL, API key, and model—also work in Cursor.