Understand Agent Hub Model Configuration

Agent Hub uses a configured large language model (LLM) to support orchestration and provide direct responses. An Agent Hub administrator configures the processing region and model that Agent Hub uses for these capabilities.

The models available for configuration depend on the selected processing region. When the administrator selects a region, Agent Hub provides a list of the models that are available for that region. Model availability can differ between regions and can change over time.

Understand the Processing Region

The processing region determines where Agent Hub performs model processing. The models available for selection depend on the selected region.

Understand Model Availability

The processing regions and models available in Agent Hub are based on Oracle Cloud Infrastructure (OCI) Generative AI availability. Model availability varies by region and can change over time.

For current information about model availability by region, see Generative AI Models by Region.

Understand the Configured Model

The model selected by the administrator becomes the configured LLM that Agent Hub uses to evaluate user requests and generate responses when no specialized agent or tool is needed.

When Agent Hub evaluates a prompt without a preselected agent, the orchestrator uses the configured LLM to help determine how to handle the request. If a specialized agent or tool isn't required, Agent Hub can use the configured LLM to provide the response directly.

The configured Agent Hub model doesn't determine which LLM an individual agent uses. Agents can use models as part of their own processing based on how those agents were developed and configured.

Why this matters

The processing region and model configuration determine which LLM Agent Hub uses for orchestration and direct responses. Understanding the relationship between region and model availability can help administrators select an available model that meets their organization's requirements.

For information about configuring these settings, see Configure the Processing Region and Model.

For information about how Agent Hub uses the configured LLM when processing user requests, see How Agent Hub Uses Large Language Models (LLMs).