Getting Started

How It Works

EnAtlas sits between your code and your LLM providers, capturing telemetry without ever touching your API keys or blocking your requests.

Install SDK or Sidecar

Add the EnAtlas SDK to your backend, or run the sidecar as a local service. Point your LLM client to it.

Every API call is captured automatically — latency, tokens, cost estimates, and status. No changes to your prompts.

View spend by provider, model, app, and workflow. Identify waste patterns and see estimated savings.

Upgrade to unlock caching, routing rules, budget guardrails, and more. Same integration, no reintegration required.