Hosted providers, your own keys, and your own GPUs behind a single endpoint at the provider's price. Routing, access control, and attribution for everything your agents call.
Change the base URL and keep your code. Hosted models, your own provider keys, and local GPUs all answer on the same API.
Catalog, usage, and limits for the whole org, broken out by agent, person, model, or day, with live request logs and the price each provider charged.
The gateway routes; the intelligence layer improves what flows through it, from model switches to cache hits to a fine-tuned model you own.
Experiential Labs is an open source AI gateway, backed by Y Combinator, built so teams can see and control every model call in one place.
Tell us what you are running and we will help you route it.