Integrated Development Environments (IDEs)

Inference Gateway is fully OpenAI-compatible, allowing you to configure it with your favorite IDEs and extensions using standard OpenAI integration settings. The following sections provide instructions for configuring popular IDEs and extensions.

VSCode

Continue.dev is a VSCode extension that supports OpenAI-compatible APIs. You can configure it to use Inference Gateway as follows:

1. Install the Continue.dev extension from the VSCode Marketplace.

2. Open the Continue.dev configuration file by running the Continue: Open Configuration command in VSCode.

3. Add a new model configuration with the Inference Gateway endpoint:

"models": [
  {
    "title": "Inference Gateway",
    "provider": "openai",
    "model": "your-model-name", // e.g., groq/deepseek-r1-distill-llama-70b
    "apiBase": "http://localhost:8080/v1",
    "apiKey": "your-api-key",
    "useLegacyCompletionsEndpoint": false
  }
]

4. Save the configuration file and restart VSCode.

Cursor

Cursor is an AI-first code editor that supports OpenAI-compatible APIs. To configure Cursor to use Inference Gateway:

1. Open Cursor and go to Settings by clicking on the gear icon in the bottom left corner or using the keyboard shortcut Cmd+, (Mac) or Ctrl+, (Windows/Linux).

2. In the settings sidebar, select AI.

3. Scroll down to find the Custom Endpoint section.

4. Enable the Use Custom Endpoint toggle.

5. Configure the following settings:

   • API Endpoint: Enter your Inference Gateway URL (e.g., http://localhost:8080/v1)
   • API Key: Enter your Inference Gateway API key
   • Model: Enter the model name you want to use (e.g., groq/deepseek-r1-distill-llama-70b)

6. Click Save to apply the changes.

Now Cursor will use your Inference Gateway for all AI operations, including code completion, chat, and other AI-powered features.