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

# FAQ

> Common questions about accounts, API keys, models, billing, and integrations in Moxus AI.

## Account and keys

<AccordionGroup>
  <Accordion title="Does Moxus AI support Google sign-in?">
    Yes. Choose the Google sign-in option on the sign-in or registration page, then complete the Google authorization flow.
  </Accordion>

  <Accordion title="How do I change my email or password?">
    Open Profile and security after signing in. You can change your login password and update the bound email address.
  </Accordion>

  <Accordion title="How do I create an API key?">
    Sign in, open API Keys, create a key, and copy the `sk-` value immediately. The full key is usually shown only once.

    <img src="https://mintcdn.com/moxus-cloud/DgOc9Y6ZmJ9TiNBD/images/api-keys-page.png?fit=max&auto=format&n=DgOc9Y6ZmJ9TiNBD&q=85&s=6558d13e388ff082201fd484d425e7f3" alt="API Keys page with the create button" width="1495" height="767" data-path="images/api-keys-page.png" />
  </Accordion>

  <Accordion title="Can I use one key for multiple projects?">
    You can, but separate keys are recommended. They make quota control, rotation, and incident response easier.
  </Accordion>

  <Accordion title="What should I do if a key is exposed?">
    Delete or disable the exposed key, create a new one, update your app, and review Usage for abnormal calls.
  </Accordion>
</AccordionGroup>

## Running code examples

<AccordionGroup>
  <Accordion title="What do I need before running a Python example?">
    `openai` is not part of the Python standard library, so install it first. Save the Python example as `image_understanding.py`, then run these commands in a terminal opened in the project folder:

    ```bash theme={null}
    python3 -m venv .venv
    source .venv/bin/activate
    python -m pip install openai
    python image_understanding.py
    ```

    This creates a project-local `.venv` environment, so dependency installation and script execution use the same Python interpreter. For later runs, first execute `source .venv/bin/activate`, then run `python image_understanding.py`.
  </Accordion>

  <Accordion title="What do I need before running a Node.js example?">
    `openai` is not a built-in Node.js module. Save the Node.js example as `image-understanding.mjs`, then run these commands in a terminal opened in the project folder:

    ```bash theme={null}
    npm install openai
    node image-understanding.mjs
    ```
  </Accordion>
</AccordionGroup>

## Models and requests

<AccordionGroup>
  <Accordion title="Why does a model call fail with model not found?">
    The model name may be wrong, unavailable, capability-mismatched, or blocked by key limits. Copy the exact model name from Model Square and check key settings.
  </Accordion>

  <Accordion title="Which Base URL should I use?">
    Use `https://moxus.cloud/v1` for OpenAI-compatible SDKs, `https://moxus.cloud` for Anthropic-compatible APIs, and `https://moxus.cloud/v1beta` for Google-compatible APIs.
  </Accordion>

  <Accordion title="Why did the request fail with 401 Unauthorized?">
    The key may be wrong, disabled, expired, or missing the `Bearer ` prefix in the `Authorization` header.
  </Accordion>

  <Accordion title="What should I do when Conversation says a parameter is deprecated or unsupported?">
    Models support different request parameters. In the Parameters section on the left side of Conversation, clear the checkbox beside the parameter named in the error, such as `Temperature` for `temperature is deprecated for this model`. An unchecked parameter is omitted from the request, so the model uses its own default behavior. You can also confirm that the field is gone in the API code snippet before sending.

    If the request still fails, clear any other parameter named in the error and try again. Changing the parameter value is not enough when the model has deprecated or does not support the field.
  </Accordion>
</AccordionGroup>

## Billing

<AccordionGroup>
  <Accordion title="Where can I see request-level cost?">
    Open Usage and check Common Logs. Each request shows model, token usage, status, cost, and failure reason when available.
  </Accordion>

  <Accordion title="Where can I see balance changes?">
    Open Billing to review balance, top-ups, redemption codes, refunds, and billing history.
  </Accordion>

  <Accordion title="Why can the same prompt cost different amounts?">
    Output length, context length, cache behavior, and model selection can vary between calls, so token usage and cost can change.
  </Accordion>
</AccordionGroup>

## Next steps

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