Getting Started
Welcome to Candela — an open-source governance platform that gives you full control and visibility over your LLM traffic. Enforce budgets, control model access, trace every request, and prove compliance.
What is Candela?
Section titled “What is Candela?”Candela sits between your applications and LLM providers as a transparent governance proxy. Every request is traced, costed, and policy-checked before it reaches the provider — and every response is logged to an immutable audit trail.
It consists of four components:
Choose Your Path
Section titled “Choose Your Path”Use candela for instant LLM governance on your machine — local models via Ollama, cloud models via ADC, all traced to SQLite with budget enforcement.
go install github.com/candelahq/candela/cmd/candela@latestDeploy candela-server on Cloud Run with BigQuery storage, Firebase auth, budget enforcement, and a dashboard UI. Developers connect via candela in Team Mode.
Drop candela-sidecar next to your application containers for zero-touch LLM governance in production.
Candela provides the governance controls enterprise teams need: per-tenant budget enforcement, immutable audit trails, model access policies, eBPF kernel-level enforcement via Tetragon, and multi-provider routing — all behind a single proxy endpoint.
Integrate Candela with your ADK agents for end-to-end distributed tracing with W3C Trace Context propagation and per-agent cost attribution.