> ## Documentation Index
> Fetch the complete documentation index at: https://docs.moxus.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Cursor

> Configure Cursor with Moxus AI's OpenAI-compatible endpoint.

Cursor can call Moxus AI through a custom OpenAI API key and Base URL. This mainly affects Chat, Composer, and Agent workflows. Tab completion behavior depends on Cursor's current version and plan rules.

## Values

| Field                    | Value                                       |
| ------------------------ | ------------------------------------------- |
| OpenAI API Key           | `sk-your-key`                               |
| Override OpenAI Base URL | `https://moxus.ai/v1`                       |
| Model name               | Copy the exact name from Models and pricing |

## Setup

<Steps>
  <Step title="Open model settings">
    In Cursor, go to Settings → Models.
  </Step>

  <Step title="Enable OpenAI API Key">
    Turn on OpenAI API Key and paste your Moxus AI `sk-` key.
  </Step>

  <Step title="Set the Base URL">
    Enable Override OpenAI Base URL and enter `https://moxus.ai/v1`.
  </Step>

  <Step title="Add or enable models">
    Search for or add the model name shown in Moxus AI's Models and pricing page.
  </Step>

  <Step title="Test a request">
    Open Chat or Composer, choose a model, and send a short test message.
  </Step>
</Steps>

## How to use it

* `Cmd/Ctrl + L`: open chat for code explanation and project questions.
* `Cmd/Ctrl + I`: open Composer for code edits.
* Agent mode: useful for multi-file tasks, but it may produce more requests and higher cost.

<Warning>
  Cursor Free may only allow Auto mode. Providing your own API key does not necessarily unlock manual model selection inside Cursor.
</Warning>

## Troubleshooting

<AccordionGroup>
  <Accordion title="Free plans can only use Auto">
    This is a Cursor plan limitation, not a Moxus AI key error. Switch to Auto or upgrade Cursor before selecting a specific model manually.
  </Accordion>

  <Accordion title="The model I want is missing">
    Add the exact model name from Models and pricing. Do not use embedding models such as `text-embedding-3-small` for coding chat.
  </Accordion>

  <Accordion title="Requests fail after setup">
    Open [Usage](/en/platform/dashboard-and-usage) and check whether the request reached Moxus AI, which key was used, and which model was called.
  </Accordion>
</AccordionGroup>

## Next steps

* [Quickstart](/en/overview/quickstart)
* [Models and pricing](/en/overview/models-and-pricing)