Skip to main content
Version: v1.1.x

Use PaletteAI Inference Launchpad with Cursor

This guide explains how to connect Cursor to a PaletteAI Inference Launchpad appliance so that a model running on the appliance serves your requests instead of a cloud provider. Cursor has no Anthropic endpoint setting, so you trust the appliance certificate, point Cursor's OpenAI override at the appliance, and confirm that requests reach it.

The console lists these steps for you. In the Connect a coding agent panel, the Cursor tab shows the values to enter. For the full list of settings, refer to Cursor Configuration.

warning

Cursor routes only Ask mode requests to a custom endpoint. Agent, Edit, and Tab remain locked to Cursor's own models. This is a limit of Cursor's bring-your-own-key support, not of the appliance, so those modes do not use the appliance even after you complete this guide.

Prerequisites​

  • Cursor installed and already working. For installation, refer to the Cursor documentation.
  • A running PaletteAI Inference Launchpad appliance with at least one model deployed and serving. To deploy a model, refer to Deploy a Model.
  • An API token for the appliance. To create one, refer to Generate an API Token, or use a token an administrator generated for you.
  • The appliance address reachable from Cursor's cloud servers, presenting a certificate those servers already trust. Cursor sends every model request from its own cloud, not from your machine. An appliance on a private network, or one that presents only a platform-issued certificate, never receives those requests, and no client-side setting changes that. For the connectivity options Cursor offers, refer to Cursor's Private Connectivity documentation.

Download the Platform CA Certificate​

When the appliance serves HTTPS with a certificate issued by its own platform Certificate Authority (CA), the connect panel offers that CA certificate as a download. Your machine does not trust the platform CA yet, so download the certificate before you configure the agent.

  1. In the console, open the Connect a coding agent panel. Select Connect Coding Agent on the homepage, select Connect a Coding Agent in the Create API token dialog that reveals a new token, or select Connect Agent in the dialog that finishes the Add client wizard.

  2. Select the tab for your agent, and then select CA certificate. The browser saves palette-ai-inference-launchpad-ca.crt to your Downloads folder.

If the panel shows no CA certificate step, the appliance does not need one. Skip this section and every later step that points at the downloaded file.

Importing the platform CA lets Cursor on your machine trust the appliance. Ask mode reaches the appliance only when the endpoint Cursor's cloud connects to presents a certificate that Cursor's cloud already trusts. If the appliance presents only a platform-issued certificate to Cursor's cloud, Ask mode cannot reach it, and no client-side step changes that.

Configure Cursor​

  1. Import the platform CA certificate into your operating system trust store, and then restart Cursor. Cursor is a graphical application, so it reads the operating system trust store rather than a shell variable. This step lets Cursor on your machine trust the appliance, so Cursor's own client-side validation of the base URL and the API key succeeds. It does not enable the model request itself, which Cursor's cloud servers send and which never consults your machine's trust store.

  2. In Cursor, open Settings > Models.

  3. In the OpenAI API Key field, enter your lpai_ token.

  4. Enable Override OpenAI Base URL, and enter your appliance address with the /v1 path appended, such as https://<appliance-host>/v1.

  5. In the model list, enable a GPT model such as gpt-5.6. The appliance seeds a gpt- alias and routes it to the model it serves, so you do not pick a backend model.

Verify the Connection​

  1. In Cursor, open a chat and set the mode to Ask.

  2. In the model picker, select the model you enabled, such as gpt-5.6.

  3. Send a test prompt.

    reply with exactly CURSOR_OK and nothing else

  4. Confirm that Cursor displays the reply CURSOR_OK.

A reply confirms that the base URL, the token, and model routing all work, and that Cursor's cloud trusted the appliance certificate.

warning

If Cursor returns We're having trouble finding the resource you requested, Cursor's cloud did not reach the appliance. Confirm that the appliance address resolves and answers from outside your network, and that its certificate is one Cursor's cloud trusts. If the appliance presents only a platform-issued certificate to Cursor's cloud, Ask mode cannot reach it, and no client-side step changes that. If Cursor answers from its own backend instead of the appliance, the model name matches an entry in Cursor's catalog. Ask an operator to add a uniquely named alias on the client's Routing tab, as described in Manage a Client's Model Access. Cursor does not list that name, so open Settings > Models, select Add model, enter the alias name, and then select it in chat.

Request Routing and Quotas​

Requests you send through the appliance are subject to the routing rule and quota configured for your client. If an operator configured a routing rule for your token, the appliance can redirect a request to a frontier model instead of a local one. Quotas apply to the client, so every API token that belongs to the client draws on the same limits. When the client exhausts a quota, the appliance returns an HTTP 429 response.

To understand what a client is and how API keys and quotas govern usage, refer to Clients and Quotas. To understand how the Tier map and the semantic router combine to pick a model for each request, refer to Routing Behavior.

Next Steps​

To look up the base URL, model name, and the Cursor modes the appliance supports, refer to Cursor Configuration. To connect a different coding tool, refer to Use PaletteAI Inference Launchpad with Claude Code, Use PaletteAI Inference Launchpad with OpenAI Codex, or Use PaletteAI Inference Launchpad with OpenCode.