Self-host the open-source Gateway in production. Pick the path that matches your stack.
The fastest way to try it locally.
docker run -p 8080:8080 \
-e VAULTPLANE_PROVIDER_OPENAI_API_KEY=sk-... \
ghcr.io/vaultplane/gateway:latestFor Kubernetes deployments.
helm repo add vaultplane https://charts.vaultplane.com
helm install gateway vaultplane/gatewayA single static binary, no runtime deps.
curl -fsSL https://get.vaultplane.com/gateway | sh
gateway serve --config ./gateway.yamlIssue scoped keys to apps and teams without sharing provider credentials.
gateway keys create --name billing-app --providers openai,anthropicPoint the Gateway at commercial and self-hosted models.
providers:
- name: openai
type: openai
api_key: ${OPENAI_API_KEY}
- name: local-llama
type: openai-compatible
base_url: http://localhost:11434/v1Export traces and metrics to any OTLP endpoint.
otel:
exporter: otlp
endpoint: http://localhost:4317Confirm the Gateway is up and routing.
curl http://localhost:8080/v1/chat/completions \
-H "Authorization: Bearer $VAULTPLANE_KEY" \
-d '{"model":"gpt-4o-mini","messages":[{"role":"user","content":"ping"}]}'Want the control plane managed? Start a Cloud workspace