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

# Providers & models

> The vendor accounts and models your agents run on.

Model Master has two related resources:

* A **Provider** is an LLM vendor account — OpenAI, Anthropic, Google, and more —
  holding your defaults and your **API key**.
* A **Model** references a provider and adds per-model settings (and cost metadata).

## Two ways to run

**Viorant** is a built-in managed provider that needs no key. It offers two tiers —
**Medium** and **Low** — and each call draws from your account's
[credit wallet](/providers/credits). This is what makes a brand-new account usable
immediately.

**Your own provider** runs models **directly from your machine** against **your** key.
Keys live in your OS credential store — Keychain on macOS, Credential Manager on
Windows — and are **never sent to Viorant's servers**. The Hub calls the provider API
itself, and these runs cost no credits.

You can mix them: different models in the same space can sit on different providers,
so it's normal to iterate on a Viorant tier and run production work on your own key.

<CardGroup cols={2}>
  <Card title="Credits & the Viorant provider" icon="coins" href="/providers/credits">
    Run on managed tiers with no setup.
  </Card>

  <Card title="Add a provider key" icon="key" href="/providers/byok">
    Bring your own key and pick a model.
  </Card>
</CardGroup>