Installation

All Candela components live in the candelahq/candela monorepo.

candela-local (Recommended Start)

The fastest way to get started — a developer proxy that manages local and cloud models with built-in observability.

go install

go install github.com/candelahq/candela/cmd/candela-local@latest

From Source

git clone https://github.com/candelahq/candela.git
cd candela
nix develop   # or ensure Go 1.26+ is installed
go run ./cmd/candela-local

Configure

Create ~/.candela.yaml:

runtime_backend: ollama
port: 8181
lm_studio_port: 1234

Solo + Cloud

Add providers and vertex_ai to also route cloud models (Gemini, Claude) through candela-local. See candela-local docs.

---

candela-server

The full backend with dashboard UI, deployed on Cloud Run.

git clone https://github.com/candelahq/candela.git
cd candela
nix develop -c go run ./cmd/candela-server

See candela-server docs for Cloud Run deployment with Terraform.

---

candela-sidecar

Lightweight production proxy for container environments.

Docker

docker pull ghcr.io/candelahq/candela-sidecar:latest

Binary

go install github.com/candelahq/candela/cmd/candela-sidecar@latest

From Source

git clone https://github.com/candelahq/candela.git
cd candela
go build -o candela-sidecar ./cmd/candela-sidecar

---

candela-desktop

Flutter desktop app for provider management and trace visualization.

Download from GitHub Releases.

---

Verify

candela-local --help

You’re ready!

Head to the Quick Start to send your first traced LLM request.