Aurora AI Gateway: A New Self-Hosted, Open-Source Option for Unified LLM Access

The AI gateway space is heating up. Today, a new open-source project called Aurora aims to simplify how developers connect their applications to multiple LLM providers. Aurora is a self-hosted, Apache-2.0 licensed gateway written in Go, designed to sit between your app and any AI provider—OpenAI, Anthropic, and more—offering a unified interface that eliminates provider-specific format headaches.

With Aurora, you can keep using the standard OpenAI or Anthropic SDKs, simply point your base_url to Aurora, and let it handle the routing. The project touts support for 14 provider types, with automatic provider discovery from environment variables. Set any provider's _API_KEY and restart—Aurora picks up the provider and its default models automatically. This is a significant convenience for developers juggling multiple API keys and want to avoid vendor lock-in.

Aurora includes both a semantic cache (vector similarity) and a response cache (exact match), which can help reduce latency and costs in production. For those needing more control, it offers custom base URL configuration, model overrides via *_MODELS environment variables, and support for Azure's API versioning and OpenRouter extras.

Getting started is quick: npm install -g iaurora and you're on your way. There is also a Docker image and Helm charts for Kubernetes deployments.

The project also offers an enterprise edition with additional features like SSO, RBAC, tenant isolation, budget enforcement, and compliance workflows. However, the core gateway is fully open source and self-hostable, making it an attractive option for teams that want to avoid the per-token markup of hosted gateway services.

Aurora joins a growing ecosystem of API gateways and proxies, but its combination of an Apache-2.0 license, OpenAI/Anthropic compatibility, and self-hosting makes it a compelling choice for developers who prioritize control and flexibility.