RPCsLive

Fastest free Solana RPC, live no-key endpoint latency

HTTP round-trip latency for getSlot at the processed commitment against every free, no-key public Solana RPC endpoint, audited every 60 seconds from 3 regions.

TL;DR. As of , Solana Labs leads rpc latency at 15 ms (p50, 24h) on Fastest free Solana RPC, live no-key endpoint latency. Source: OpenChainBench, https://openchainbench.com/benchmarks/solana-rpc.

Companion page

Comparing free RPC endpoints across every chain we measure? Open the cross-chain RPC matrix →

Solana is the first non-EVM member of the RPC latency cluster, and the chain where the free-RPC question is asked most and answered worst: most "best Solana RPC" roundups list endpoints that key-gated or died years ago. This page probes the 5 endpoints that actually answer keyless today, including the Solana Labs official api.mainnet-beta.solana.com that every tutorial pastes, with the same rules as the 20 EVM chains in the cluster: one identical probe (getSlot at the processed commitment) every 60 seconds from us-east, eu-west and Singapore, stale-slot detection against the cross-provider tip, and result classification so a fast error never ranks as a fast answer. The candidates that failed the keyless audit are listed with their exact refusals: dRPC moved Solana to paid tiers, Ankr returns 403 without a key, OnFinality's shared public quota is permanently exhausted.

Methodology

Per-chain member of the RPC latency cluster and its first non-EVM chain. We measure round-trip latency of a single identical JSON-RPC call (getSlot at the processed commitment) against every no-key public Solana endpoint that sustains continuous probing, 5 providers, every 60 seconds, from us-east, eu-west and Singapore. Responses are classified (ok / http_err / jsonrpc_err / stale / timeout) with staleness measured in slots against the cross-provider tip. The archive-depth audit run on EVM chains has no keyless Solana equivalent and is not performed. The cross-chain view lives on the parent rpc-capabilities benchmark; this page is the Solana-scoped answer with per-region breakdowns as a first-class dimension.

Frequently asked

What is the fastest free Solana RPC right now?

Solana Labs currently leads at 15 ms (getSlot p50 over the last 24h), measured against 5 no-key providers probed every 60 seconds from us-east, eu-west and Singapore. The leaderboard re-sorts continuously against fresh Prometheus samples; use the region tabs to see the leader from the origin closest to your deployment.

Which Solana RPCs work without an API key?

The 5 providers on this page: Solana Labs (api.mainnet-beta.solana.com), PublicNode, Lava, LeoRPC (publicly documented FREE key) and Solana Vibe Station. Every endpoint was live-verified before inclusion. The famous names that do not work keyless anymore are listed in the methodology with their exact refusals: dRPC moved Solana to paid tiers, Ankr 403s, OnFinality's public quota is exhausted, Helius and Shyft are key-gated.

Is api.mainnet-beta.solana.com good enough for production?

For development, demos and light read workloads, yes: Solana Labs documents roughly 100 requests per 10 seconds per IP. For production trading, indexing or anything latency-sensitive, dedicated providers exist for a reason; this page measures the free tier so you know exactly what the default gets you before you decide.

How is Solana RPC latency measured here?

One identical JSON-RPC POST (getSlot at the processed commitment) every 60 seconds against each provider from each of 3 regions. Wall-clock round-trip is recorded at millisecond precision; p50/p90/p99 come from Prometheus quantile_over_time over 24 hours. Responses are classified so an endpoint stuck behind the cross-provider slot tip or erroring behind HTTP 200 is never ranked as fastest. The harness is open source.

Why is there no archive-depth column like the EVM pages?

The EVM cluster probes eth_getBalance at historical block heights to expose which free endpoints serve archive state. Public Solana endpoints expose no comparable keyless historical-state query, so the audit is disclosed as not applicable rather than approximated.

Source code github.com/ChainBench/OpenChainBench/tree/main/harnesses/rpc-capabilities