Skip to main content
Version: v1.0.x

PaletteAI Inference Launchpad Cursor Configuration

This page lists the configuration values Cursor uses to connect to a PaletteAI Inference Launchpad appliance. For the steps to set them, refer to Use PaletteAI Inference Launchpad with Cursor.

Settings​

Set the following in Cursor under Settings > Models.

SettingDescriptionExample value
Override OpenAI Base URLThe PaletteAI Inference Launchpad inference endpoint. Use the appliance address with the /v1 path appended.https://amd.spectrocloud.com:8443/v1
OpenAI API KeyThe API token generated in the console. It begins with lpai_.lpai_YOUR_TOKEN
Add modelThe unique alias name the appliance serves. Enter the alias, not the backend model id.launchpad-glm52

Endpoint URL​

Cursor uses the OpenAI-compatible API, which the gateway serves at /v1. Set Override OpenAI Base URL to your appliance's address, the same host you use to reach the console, with /v1 appended. This differs from Claude Code, which uses the Anthropic Messages API and adds the path itself.

The appliance must have a valid, publicly trusted TLS certificate on a DNS name. Cursor sends the request from its own cloud servers rather than from your machine, so it does not accept a self-signed certificate and offers no way to skip certificate verification.

Model Name and Aliases​

Cursor decides where to route a request by the model name. If the name matches a model in Cursor's built-in catalog, such as glm-5.2 or gpt-4o, Cursor routes the request to its own backend, and the appliance receives nothing. To reach the appliance, an operator must serve the model under a unique alias name that does not appear in Cursor's catalog.

An operator creates the alias on the appliance, which maps it to a model the appliance already serves, such as zai-org/GLM-5.2. Both the alias and the ids of the models the appliance serves appear in the console model list and in the appliance's /v1/models API response.

Supported Modes​

Cursor modeSupportedNotes
Ask (chat)YesRoutes to the appliance through the model alias.
AgentNoLocked to Cursor's own models for bring-your-own-key setups.
EditNoLocked to Cursor's own models for bring-your-own-key setups.
TabNoLocked to Cursor's own models for bring-your-own-key setups.

The unsupported modes are a limitation of Cursor's bring-your-own-key support, not of the appliance.

Token Quotas​

If the token's quota is exhausted, the appliance returns an HTTP 429 response and Cursor surfaces the error.

Resources​