AI is the single gateway every other part of the mesh uses to reach large language models, and you can use it directly too. It gives you one consistent endpoint for chat completions, streaming responses, and text embeddings across multiple providers, so you can switch models without rewriting anything.
Your provider keys are held by the gateway and never exposed to the components calling through it, and every request is metered by tokens and request count so usage stays transparent. You get dependable, provider-agnostic access to modern AI with central key management and clear accounting, instead of wiring each tool to a provider by hand.
No capabilities accessible with your current permissions. Visit the hub to browse available products.