Model providers
bookistudios AI runs agents on LLM providers through OpenAI-compatible endpoints. A provider is configured with a base URL and an API key; models are addressed by their provider-specific model or endpoint ID.
Any provider that exposes an OpenAI-compatible chat or responses API can be wired in — the pages in this section document the integrations that are known to work well with bookistudios AI.
Supported providers
How provider configuration works
Provider settings live in the config file under the model provider section:
model_provider:
base_url: https://...
api_key_env: PROVIDER_API_KEY
model: provider-model-idThe API key is read from the environment variable named by api_key_env — keys are never stored in the config file itself. See Configuration for the full configuration reference.
Adding a new provider
Most OpenAI-compatible services work out of the box: set base_url to the service endpoint, point api_key_env at your key, and use the service’s model identifier. If you have a provider you would like documented here, the integration guide covers the exact wiring.