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

# Models & Providers

> The catalog of selectable models and what each one is good for.

Custral works with multiple AI models from OpenAI and Anthropic. The same chat or agent can run on any of them. Pick the model that fits the job, and switch at any time without losing your setup.

## Catalog

| Model             | Provider  | Context window | Tools | Vision | Tier      |
| ----------------- | --------- | -------------- | ----- | ------ | --------- |
| GPT-4o            | OpenAI    | 128K           | ✓     | ✓      | balanced  |
| GPT-4o mini       | OpenAI    | 128K           | ✓     | ✓      | fast      |
| o1-mini           | OpenAI    | 128K           | ✗     | ✗      | reasoning |
| Claude Opus 4.7   | Anthropic | 1M             | ✓     | ✓      | reasoning |
| Claude Sonnet 4.6 | Anthropic | 200K           | ✓     | ✓      | balanced  |
| Claude Haiku 4.5  | Anthropic | 200K           | ✓     | ✓      | fast      |

**Context window** is how much text the model can consider at once. **Tools** indicates whether the model can take actions (read records, send email, search the web). **Vision** indicates whether it can read images.

The default model for a new workspace is GPT-4o.

## Tiers

Each model belongs to a tier that signals its trade-off between speed and depth:

<CardGroup cols={3}>
  <Card title="Fast" icon="bolt">
    Quick, low-cost responses. Best for high-volume property auto-fill and simple chat. (GPT-4o mini, Claude Haiku 4.5)
  </Card>

  <Card title="Balanced" icon="scale-balanced">
    The safe default for most chat and agent work. (GPT-4o, Claude Sonnet 4.6)
  </Card>

  <Card title="Reasoning" icon="brain">
    Deeper reasoning for complex chat and schema design. (Claude Opus 4.7, o1-mini)
  </Card>
</CardGroup>

## Which model runs

When you send a message, Custral picks the model to use in this order:

<Steps>
  <Step title="The model you pick in chat">
    A model chosen directly in the chat takes precedence over everything else.
  </Step>

  <Step title="The conversation's model">
    Once a conversation starts, its model sticks for the life of that conversation.
  </Step>

  <Step title="The agent's model">
    A chat agent can declare a preferred model. Set it to **Auto** to defer to the workspace default.
  </Step>

  <Step title="The workspace default">
    Set on the workspace AI settings page, this applies when nothing more specific is chosen.
  </Step>

  <Step title="Fallback">
    If no preference applies anywhere, Custral uses GPT-4o.
  </Step>
</Steps>

Some background tasks (like generating a short description) always use a fast, low-cost model regardless of the choice above.


## Related topics

- [AI](/ai/overview.md)
- [Get started with AI](/ai/getting-started.md)
- [Chat Agents](/ai/chat-agents/overview.md)
- [Billing & Credits](/start/custral/billing.md)
- [Data Models](/data/models.md)