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

# Configuring AI providers

AI providers are the model connections that your agents use for inference. Each provider gives you access to one or more models that you can select when creating or editing an agent.

To view and manage your AI providers, select <Icon icon="brain-circuit" iconType="light" /> **AI Providers** from the left-hand navigation.

## Anaconda built-in provider

Agent Studio includes an **Anaconda** provider with hosted models accessible through your Anaconda account.

Models available through the Anaconda provider include:

* Claude Sonnet 4.6 (default), Claude Haiku 4.5
* GPT-OSS 120B
* Nova Pro, Nova Lite, Nova Micro

## Anaconda Desktop provider

Agent Studio includes an **Anaconda Desktop** provider that connects to models running locally in Anaconda Desktop's model server.

To use a local model:

1. Start a model server in Anaconda Desktop.
2. Return to Agent Studio. The running model appears automatically in the Anaconda Desktop provider's model list.

The model is then available as an option when creating or editing agents. Your agents use the local model for inference without external API calls.

## Adding a provider

Agent Studio supports direct connections to OpenAI, Anthropic, Google, and Groq with simplified setup, as well as custom providers for any endpoint that uses a compatible protocol. Once connected, the provider's models become available when creating or editing agents.

<Tabs>
  <Tab title="Supported providers">
    1. Select <Icon icon="brain-circuit" iconType="light" /> **AI Providers** from the left-hand navigation.
    2. Click <Icon icon="plus" iconType="light" /> **Add**.
    3. Select a provider.
    4. Enter your API key.

    <Tip>
      Agent Studio validates the connection and discovers the models available to your account.
    </Tip>
  </Tab>

  <Tab title="Custom providers">
    1. Select <Icon icon="brain-circuit" iconType="light" /> **AI Providers** from the left-hand navigation.
    2. Click <Icon icon="plus" iconType="light" /> **Add**.
    3. Select **Custom Provider**.
    4. Enter a name for the provider.
    5. Select the protocol your endpoint uses.
    6. Enter the endpoint URL for the provider.
    7. If necessary, enter your API key.

    <Tip>
      Agent Studio validates the connection and lists the available models.
    </Tip>
  </Tab>
</Tabs>

## Editing and deleting providers

1. Select <Icon icon="brain-circuit" iconType="light" /> **AI Providers** from the left-hand navigation.
2. Click <Icon icon="ellipsis" iconType="light" /> on the provider's row.
3. Select <Icon icon="pencil" iconType="light" /> **Edit** to modify the provider's configuration, or select <Icon icon="trash" iconType="light" /> **Delete** to remove it.

<Note>
  Deleting a provider removes access to its models. Any agents using that provider will need to be reconfigured with a different AI provider.

  ***

  The Anaconda and Anaconda Desktop providers cannot be edited or removed.
</Note>