Self‑hosted gateway unifying Anthropic, OpenAI, Google, and Bedrock APIs
MLPal Gateway lets you run a single local endpoint that routes requests to multiple large‑language‑model providers, handling authentication, rate‑limiting, and per‑key budgeting. It translates a unified Anthropic‑style request into each provider's native API, preserving features like caching and tool use. Designed for developers and teams who need a consistent interface and transparent cost tracking across providers. Compared to ad‑hoc scripts, it offers a polished, production‑ready solution with observability and admin UI.
View on GitHub →mlpalOld/mlpal-gateway