LLM Providers

• 1: Anthropic
• 2: AWS Bedrock
• 3: Azure Foundry
• 4: Google Gemini
• 5: OpenAI

Overview

Atolio makes it easy to securely configure and manage access to commercial LLMs of your choice. Atolio supports multiple LLM models across different providers, allowing you to select the best model for your organization’s needs.

High-level Configuration Workflow

To configure an LLM provider, navigate to the administration panel at your Atolio deployment URL (e.g., https://search.example.com/admin) and select LLM Provider from the menu.

Atolio uses different models for different kinds of work, and supports a range of providers. Models are updated as improved ones become available. Per provider, the following is currently made available within Atolio. Select Configure for your provider’s setup steps:

Provider	Supported Models	Configuration
Anthropic	Sonnet 5.5 Haiku 4.5	Configure
AWS Bedrock	GPT-5.6 Sol GPT-5.6 Luna Sonnet 5.5 Haiku 4.5	Configure
Azure Foundry	GPT-5.6 Sol GPT-5.6 Luna	Configure
Google Gemini	Gemini Flash 3.6	Configure
OpenAI	GPT-5.6 Sol GPT-5.6 Luna	Configure

Configuring a Provider

Most providers require an API key. AWS Bedrock is the exception: it uses the IAM role of your AWS deployment instead. To enable a provider:

1. Open Atolio and head to the Admin panel
2. Select LLM Provider from the left-pane menu
3. Select the LLM Provider and follow the steps on its page, linked in the table above.

Multiple LLM Models

You can easily configure and manage multiple LLM providers within a single Atolio instance. While only one provider is active at a time, this flexibility enables you to test new providers, compare their performance, and maintain fallback options in case of credential expiration or service issues.

Key capabilities:

• Multiple Provider Support: Add as many LLM provider configurations as you need. Anthropic and AWS Bedrock share the same namespace, as do Azure and Azure OpenAI.
• Default Management: Easily switch between providers by setting any as the default with a single click. For existing systems, the first configured provider becomes the implicit default. When additional providers are added, the previous default is preserved.
• Protected Defaults: The currently active default provider cannot be removed to ensure continuity of service.

Changing Providers

As described in the Multiple LLM Models section, you can freely switch between configured LLM providers by setting a different provider as the default. This allows you to test new providers or change providers without requiring system reconfiguration.

1 - Anthropic

Atolio connects to Anthropic’s Claude models directly with an Anthropic API key.

Steps

1. Create an Anthropic API key.
2. Add the key to Atolio.

Create the API Key

1. Sign in to the Claude Developer Platform at https://platform.claude.com/.
2. Create a new API key and copy it.

Configure Atolio

1. In Atolio, open the Admin panel and select LLM Provider.
2. Choose Anthropic and enter the API key.

Note: Anthropic and AWS Bedrock share the same namespace. See Multiple LLM Models.

2 - AWS Bedrock

Atolio can use models hosted on Amazon Bedrock. Bedrock does not use an API key: Atolio must be deployed on AWS, and it calls Bedrock as the Marvin service role marvin_svc_act. The Atolio Terraform deployment already grants this role permission to invoke Bedrock models.

Steps

1. Enable the model in Bedrock.
2. For AWS Marketplace models only, grant Marketplace permissions.
3. Enable AWS Bedrock in Atolio.

Enable the Model

In the AWS console, open the Amazon Bedrock playground and invoke the selected model once. This enables the model for your account and accepts its terms of use.

Grant Marketplace Permissions

Skip this step unless the selected model is an AWS Marketplace model, for example a third-party model with Marketplace billing.

Append the following statement to the existing marvin_svc_act role:

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "aws-marketplace:ViewSubscriptions",
                "aws-marketplace:Subscribe"
            ],
            "Resource": "*"
        }
    ]
}

Configure Atolio

1. In Atolio, open the Admin panel and select LLM Provider.
2. Choose AWS Bedrock, select the model family (Anthropic or OpenAI), and enable it. The region is selected automatically from the region Atolio is deployed in.

For more details, see the Amazon Bedrock User Guide.

Note: Anthropic and AWS Bedrock share the same namespace. See Multiple LLM Models.

3 - Azure Foundry

Atolio can use OpenAI models that you deploy in your own Azure AI Foundry resource, so requests stay within your Azure tenant.

Note: In the Atolio Admin panel, this provider is currently listed as Azure OpenAI.

Steps

1. Create an Azure AI Foundry resource.
2. Deploy both models.
3. Add the endpoint and key to Atolio.

Create the Foundry Resource

1. Open the Azure AI Foundry Portal at https://ai.azure.com/ and create a new Azure AI Foundry resource.
2. Choose the subscription, resource group, region, and resource name, then create the resource.

Deploy the Models

Atolio needs both models deployed in the same resource.

1. From the service home page, open Deployments from the left-hand menu.
2. Create a base model deployment for gpt-5.6-sol.
3. Create a second base model deployment for gpt-5.6-luna.
4. On the model deployment page, under Endpoint Settings, copy the API key and the base Endpoint URL.

Configure Atolio

1. In Atolio, open the Admin panel and select LLM Provider.
2. Choose Azure OpenAI (the Azure Foundry provider) and enter the Endpoint URL and API key.

Note: Azure and Azure OpenAI share the same namespace. See Multiple LLM Models.

4 - Google Gemini

Atolio connects to Google’s Gemini models with an API key from Google AI Studio.

Steps

1. Create a Google AI Studio API key.
2. Add the key to Atolio.

Create the API Key

1. Sign in to Google AI Studio at https://aistudio.google.com.
2. Create a new project and API key, and copy the key.

Configure Atolio

1. In Atolio, open the Admin panel and select LLM Provider.
2. Choose Google and enter the API key.

5 - OpenAI

Atolio connects to OpenAI’s GPT models directly with an OpenAI API key. To use OpenAI models hosted in your own Azure tenant, see Azure Foundry.

Steps

1. Create an OpenAI secret key.
2. Add the key to Atolio.

Create the Secret Key

1. Open the API keys page at https://platform.openai.com/api-keys.
2. Create a new secret key and copy it.

Configure Atolio

1. In Atolio, open the Admin panel and select LLM Provider.
2. Choose OpenAI and enter the secret key.