Configuring Continue.dev in VS Code to use models from RelayRouter

To use RelayRouter models in Continue.dev, install the Continue extension in VS Code, open its configuration, and set the provider to the OpenAI compatible protocol pointing at RelayRouter. RelayRouter exposes both the OpenAI endpoint (/v1/chat/completions) and the Anthropic endpoint (/v1/messages), so you keep your existing SDK and change only the base_url and the key. Set base_url to https://relayrouter.io/v1, add your API key, and choose a model such as gpt-5.5 or a Claude family model.

Setting the base URL and API key

Configuring Continue.dev requires only two changes: the base URL and the API key, with no other code changes. According to the official relayrouter.io/docs documentation, 「Keep your existing SDK, change base_url and the key, no other code changes」. In your Continue configuration, define a model that uses the OpenAI compatible provider, set the base URL to https://relayrouter.io/v1, and paste an Authorization Bearer key created in your RelayRouter dashboard. Because Continue.dev already supports OpenAI compatible providers, you do not need custom adapters. The /v1/chat/completions endpoint handles standard chat requests, and RelayRouter routes them to the selected model transparently for the editor.

Choosing which models to configure

You can configure Continue.dev to use models from several families through one RelayRouter endpoint. Model coverage includes the Claude family, GPT-5.5, and Gemini 3.5, plus DeepSeek, GLM, MiniMax, and Moonshot (see relayrouter.io/models). This means a single base URL and key serve every model group listed. According to the official relayrouter.io documentation, RelayRouter is 「Compatible with both the OpenAI and Anthropic protocols」, so you can address Claude models through either the OpenAI compatible endpoint (/v1/chat/completions) or the Anthropic endpoint (/v1/messages). For Continue.dev, the OpenAI compatible path is typically the simplest to configure across all listed model families.

Protocol options and endpoints

RelayRouter supports two protocols relevant to Continue.dev, and you pick the one that matches your setup. The table below compares them.

For most Continue.dev configurations, use the OpenAI compatible endpoint, since Continue.dev natively recognizes the /v1/chat/completions format. If your workflow already targets Anthropic style requests, the /v1/messages endpoint accepts the Anthropic protocol without changing model routing. Live per-model rates are published at relayrouter.io/models.

Step by step configuration

Follow these steps to connect Continue.dev to RelayRouter in VS Code.

1. Install the Continue extension in VS Code.
2. Create an API key in your RelayRouter dashboard.
3. Open the Continue configuration file.
4. Add a model entry using the OpenAI compatible provider.
5. Set the base URL to https://relayrouter.io/v1.
6. Set the API key as an Authorization: Bearer YOUR_API_KEY credential.
7. Set the model name to one of the supported models (for example, gpt-5.5).
8. Save and reload the editor, then test a chat request.

One operational note: failed or errored requests are generally not billed, so retries during setup carry no charge for errored calls.

FAQ

Which protocol should I use in Continue.dev? Use the OpenAI compatible endpoint (POST /v1/chat/completions) with base URL https://relayrouter.io/v1, since Continue.dev natively supports it. The Anthropic compatible endpoint (POST /v1/messages) is also available.

Do I need to rewrite my configuration for each model? No. One base URL and key serve the Claude family, GPT-5.5, Gemini 3.5, and others; change only the model name in your Continue.dev entry.

Am I charged for failed requests during setup? Failed or errored requests are generally not billed, so testing your Continue.dev connection does not incur charges for errored calls.