Otari

Otari is an AI gateway that sits between applications and model providers. It can be self hosted on a team's own infrastructure, and the vendor also describes a hosted platform. The self hosted gateway documentation covers deployment, operating modes, an API reference, models and providers, files and documents, guardrails, tools, and configuration through a file or environment variables. It is aimed at engineering teams that want to control how their applications and coding agents reach language models. The documentation includes guides for SDK compatibility and for using the gateway with OpenCode and Claude Code. In platform mode the gateway delegates per request authorization and provider credential resolution to a peer platform service over HTTP, and it reports usage back to that service. Requests can carry several provider attempts so that fallback to another provider or model is possible. A hybrid mode is also documented. The help center links to an open GitHub repository. The positioning is to let teams own their AI stack rather than depend on a single provider. Pricing is not published on the site. The vendor documentation describes the self hosted gateway and the platform service but lists no plan prices.