Model Selection
At the top of the configuration screen, you can select the Provider and Model that power your Agent’s intelligence. This choice defines both the LLM used for reasoning and the billing behavior for API usage.
How Providers Work
There are two main ways to connect LLMs to your Agents:
Option 1 — Using Zaia Provider
When you select Zaia as the provider, your Agent uses the models available inside the platform (such as claude-sonnet-4-5 or gpt-4o-mini). In this mode:
You consume Zaia platform credits for each Agent execution.
All authentication, API connections, and billing are handled automatically by Zaia.
No API key configuration is required.
This option is ideal for most users who want a fully managed experience with instant access to multiple LLMs.
Option 2 — Adding a Custom Provider
You can also connect your own LLM provider directly, bypassing Zaia’s managed credits. To do so, go to:
Platform → Settings → Providers → Configure ProviderFrom this panel, you can:
Enter a custom provider name (e.g., “OpenAI Personal” or “Anthropic Enterprise”).
Select the provider type (Anthropic, OpenAI, or Google AI).
Paste your own API key from the provider’s dashboard.
Optionally add a short internal description.
Once saved, your provider will appear in the model selection dropdown. When your Agent uses this provider:
It consumes your own API credits, not Zaia’s.
You are billed directly by the external provider.
The connection remains secure and isolated under your account.
Note: Zaia only stores the token in an encrypted format and does not transmit it externally except when invoking the selected model for your Agent.
Last updated