Available on: Starter+ | Role: Organization Owner only | Status: Generally Available

​
Overview

The model connection wizard guides you through selecting a provider, choosing a model, and configuring credentials. Once validated, the model becomes available for use across your organization.

​
Add a Custom Model

1

Navigate to Models

Click Models in the left sidebar, then click the Add Model button in the top-right corner.
If the Add Model button shows a lock icon, your plan doesn’t support custom models. Upgrade to Starter or higher to unlock this feature.
2

Select a Provider

Use the provider filter dropdown to narrow down by provider (OpenAI, Anthropic, Azure, etc.). You can also filter by capability (Chat, Vision, Embeddings, etc.).
3

Choose a Model

Browse the filtered model list on the left and click on the model you want to register. The right panel shows configuration fields for the selected model.
4

Configure Credentials

Fill in the required fields for your chosen provider:Toggle Show Advanced Configuration for additional parameters like custom headers or timeout settings.
You can switch between Form View and JSON View using the toggle at the top of the configuration panel.
5

Connect and Validate

Click Connect the model. The system will:
  1. Register the model with the Horus AI engine
  2. Run a health check to verify connectivity
  3. Save the model to your organization
If the health check fails, you’ll see an error message with details. Double-check your API key and endpoint URL.

Add a model now →

Open the model connection wizard and register your AI model

​
After Adding a Model

Once your model is successfully registered:
  • It appears in the Models Overview page with its provider icon and capabilities
  • Admins can assign it to teamspaces as a conversation, vision, tools, or moderation model
  • It becomes available for selection in AI Box configurations