The hosted dashboard turns the binary's telemetry into provider health history, per-method cost analytics, failover and error logs, and alerts — across every provider, in one place. You keep running the binary with your own keys; we host the visibility on top.
The binary keeps talking to your providers directly. It batches and posts lightweight telemetry — method, provider, status, latency, estimated cost — to the dashboard. The dashboard never proxies your RPC traffic and never sees your provider credentials.
# rpc-plane.toml [reporting] endpoint = "https://app.rpcplane.dev/api/ingest" api_key = "rp_live_xxxx" flush_interval_ms = 5000
One pane of glass across every provider — not five separate consoles that each only see their own slice.
Health scores, slot height, and slot drift over time, per provider — so you can prove which provider degraded and when.
Estimated spend broken down by method and by provider. Find the expensive call — usually not the obvious one. See the rate data →
Filter recent traffic by method, provider, status, and time. Bucketed counts, error rate, and p50/p95/p99 latency at a glance.
Every failover the binary made — from which provider, to which, and why. The audit trail for "what happened at 2am."
A dedicated, low-noise stream of errors by code, method, and provider — so a provider's bad rollout is obvious immediately.
Hit/miss rates by method, so you can see exactly which reads are being served from cache versus hitting a provider.
When cross-provider validation ships in the binary, every detected disagreement between providers lands here with the resolution.
Separate keys per environment tag traffic as dev, staging, or prod — so production metrics never get muddied by test runs.
History beyond what Prometheus on a single box keeps — go back weeks to investigate a regression or reconcile a bill.
Set team-wide thresholds and route alerts to the channels you already watch. The dashboard evaluates triggers continuously and fans matching incidents out within minutes.
Invite teammates with owner / admin / member roles. Everything — telemetry, alerts, audit log — is scoped to your team.
Claim your company email domain so colleagues discover and join the right workspace automatically — approve-on-request or auto-join.
Model each service as an app with its own keys and per-app policy: rate limit, budget cap, IP/origin allowlist, and method blocklist.
Multiple keys per app: create a new key, flip the binary config, revoke the old one — no downtime, retrievable any time from the dashboard.
Every mutation — key created, member added, policy changed — is recorded with actor and timestamp.
Enterprise SSO via SAML / Okta / Azure AD is available for teams that need it.
The dashboard is free while in private beta. At general availability, paid plans will differ on telemetry retention, team seats, and number of apps — the binary always stays free. Final tiers and prices will be published before billing turns on.
Tier structure shown for planning. Pricing is not yet active — the beta is free.
The proxy. Routing, failover, health scoring. Your keys, your machine.
Hosted observability, cost analytics, and alerts.
We run the proxy too. Configure providers in the UI, zero ops.
No. The binary is fully standalone and free. The dashboard is an optional hosted layer for history, cost analytics, and alerting.
No. The binary talks to providers directly; it sends only telemetry to the dashboard. The dashboard never proxies RPC traffic and never holds your credentials.
Sign in with GitHub or Google, copy your app key, and add a [reporting] block to rpc-plane.toml. Nothing else about the binary changes.
The binary is source-available (ELv2). The hosted dashboard and managed cloud are proprietary — you log in rather than self-host.
Free during the private beta. Paid plans with longer retention, more seats, and more apps are planned for general availability.
Request beta access below, or sign in directly at app.rpcplane.dev.
Drop your email for a private-beta invite, or sign in now with GitHub or Google.