Model Providers

A model provider is the per-tenant LLM configuration agents run on: the provider type, its base URL, the models it exposes, a default model, and a max-token cap.

Version one: Groq

Version one runs agents on the Groq provider. When an active Groq model provider is configured:

• Agent create/update validates the agent's model against the provider's model list.
• Runs use the provider's max-token cap for model calls.

When no provider row is configured, the platform falls back to its static defaults — useful for development, but configure a real provider for production tenants.

Configuration

Field	Purpose
Name	Display name.
Provider	The provider type (Groq in v1; the model supports Anthropic, OpenAI, Google, Ollama, DeepSeek for future versions).
Base URL	The provider's API endpoint.
Models / default model	The models agents may select, and the default.
Max tokens	Output-token cap applied to model calls.

API keys

The provider's API key is never stored in Agentfleet's database. The platform reads it from secure configuration (a key vault in deployed environments), and the provider record holds only a reference. Keys are separated per environment — staging and production use different keys.

Lifecycle

Providers are Active, Inactive, Disallowed, or Deleted. Only an active provider participates in model resolution and agent-model validation.