Transom

LLM gateway

Also called: AI gateway, model proxy, LLM proxy

A proxy that sits between applications and one or more model providers, adding routing, key management, rate limiting, logging, and cost controls.

An LLM gateway is a single endpoint that applications call instead of talking to OpenAI, Anthropic, or a self-hosted model directly. It holds the provider credentials, decides which backend serves each request, enforces quotas, and records traffic. Open-source examples include LiteLLM; every major cloud now offers a managed one.

Why it matters

A gateway concentrates value: it stores every upstream API key and often logs full prompts and responses. A gateway exposed to the internet without authentication hands an attacker free model usage billed to the operator, and potentially the contents of the logs and the upstream keys. Exposed gateways are one of the two things this project's map tracks, because they are common, easy to misconfigure, and expensive when left open.

Exposure map

Live counts of instances showing this pattern will appear here once the exposure map is collecting data.

Related terms

Added 2026-09-02. Last reviewed 2026-09-02. Definitions in this space are evolving; entries are dated so revisions stay legible.