With the bundled model, your questions and supplied text are processed on your computer. You do not need an AI provider account or API key for local inference.
The engine starts when needed and stops after an idle period. It uses system memory while working; performance depends on your hardware and available resources.
Ask Scribe is separate from the Smart Content Assistant, which handles lesson generation, graphics, research, and fact checking through connected AI providers.
Local processing describes the bundled model. If IT configures a remote Ollama server, that server receives the questions and context sent to it.