Stop babysitting RPC providers.
eRPC sits in front of every upstream you have — managed providers, your own node, anything — and handles failover, retries, response caching, and multi-chain routing automatically. One URL pattern, every EVM chain, zero provider lock-in.
Maintained by the team at
Goldsky# erpc.yaml — one endpoint, every provider
projects:
- id: main
upstreams:
- endpoint: ${RPC_PROVIDER_1} # your primary provider
- endpoint: ${RPC_PROVIDER_2} # a fallback provider
- endpoint: http://localhost:8545 # your own node
networks:
- chainId: 1
failsafe:
retry:
maxAttempts: 3
hedge:
delay: 200ms
consensus:
requiredParticipants: 2One proxy in front of every RPC provider
Your apps and indexers send requests to a single eRPC endpoint. eRPC serves from cache when it can, and otherwise routes to the healthiest upstream — handling rate limits, failover, and request splitting automatically.
Enterprise-grade observability
Real-time Grafana dashboards give you full visibility into usage, performance, and health — across every chain, method, and upstream.
Everything a production RPC layer needs
eRPC bundles the reliability, caching, and routing logic you would otherwise stitch together yourself — into a single configurable proxy.
Fault-tolerant proxy
Retries, circuit breakers, failovers, and hedged requests route every call to the fastest healthy upstream.
Reorg-aware caching
A persistent cache eliminates duplicate reads and stays consistent across chain reorgs.
Automatic method routing
eRPC tracks which provider supports which eth_* method and routes each request automatically.
Consensus & integrity
Compare results across providers, enforce consensus, and penalize nodes that disagree or return stale data.
Configurable rate limits
Per-provider hourly and daily quotas cap usage and keep you inside your compute-unit budgets.
Selection policies
Serve cheap upstreams until they breach error-rate or block-lag thresholds, then fail over to premium nodes.
Unified errors
Normalized error codes and messages across every provider, with integrity checks built in.
Flexible authentication
Secrets, JWT, SIWE, and basic auth secure every endpoint for trusted access.
Observability
Track throughput, errors, and latency per upstream from one real-time dashboard.
Proven in production
Teams running real workloads cut costs and stabilized their infrastructure by putting eRPC in front of their providers.
eRPC is the RPC layer behind Cloud Edge. It's how we serve fast, consistent blockchain data across hundreds of chains for some of the largest teams in crypto.
“We used eRPC to cut RPC requests by two-thirds, improving reliability and operational efficiency across our chains.”
“eRPC consolidated our RPC across 18+ chains with smarter caching and multiplexing — one reliable layer for everything.”
“Migrating from our in-house JSON-RPC system to eRPC removed infrastructure bottlenecks and stabilized CPU usage across our multi-chain setup.”
“eRPC keeps our public testnet and devnet traffic flowing across three continents, at scale.”
Every EVM chain, in one line
Self-hosting eRPC? Add the Cloud Edge provider and a single line auto-imports every chain Goldsky serves — no per-network config.
# erpc.yaml — every EVM chain, one line
projects:
- id: main
upstreams:
# Cloud Edge provider — auto-imports every chain it serves
- endpoint: goldsky://${GOLDSKY_API_KEY}