FAQ from LLM Gateway
What is LLM Gateway?
LLM Gateway is an open, extensible control plane for generative AI infrastructure—empowering teams to route, govern, monitor, and optimize LLM traffic across heterogeneous backends without rewriting application logic.
How do I integrate it?
Point your existing OpenAI-compatible client to `https://api.llmgateway.io/v1`, authenticate with your LLM Gateway API key, and let the gateway handle provider dispatch, retry logic, logging, and cost attribution—transparently and at scale.
What sets LLM Gateway apart from OpenRouter?
LLM Gateway prioritizes operational autonomy: fully self-hostable, MIT-licensed, with no enforced vendor markup—even on paid tiers. Its analytics go deeper (per-prompt cost allocation, response-time percentiles, model-level drift detection), and its routing engine supports custom policies (e.g., “route coding queries to Claude-3.5-sonnet unless latency > 800ms, then fallback to Groq”).