Base RPC endpoints: free public URLs and private API-key providers by latency
HTTP round-trip latency for eth_getBlockByNumber against every free, no-key public Base RPC endpoint (probed every 60 seconds) and, under the Access selector (Private), against the API-key endpoints of Alchemy, Chainstack, GetBlock, Quicknode and Dwellir (every 120 seconds), from 3 regions. The two cohorts are ranked separately, never against each other.
TL;DR. As of , PublicNode has the lowest median latency of the 5 free public Base RPC endpoints measured, 76 ms (p50, 24h, 3 regions). Source: OpenChainBench, https://openchainbench.com/benchmarks/base-rpc. Chain ID 8453. 5 public endpoints, no key: base-rpc.publicnode.com, base.drpc.org, gateway.tenderly.co, mainnet.base.org, base-pokt.nodies.app. Full URLs with their 24h median are listed below; Alchemy, Chainstack, GetBlock, Quicknode, Dwellir, BlockMachine (private, API key) are ranked apart under the Private tab.
Companion page
Comparing free RPC endpoints across every chain we measure? Open the cross-chain RPC matrix →
Base offers the cleanest official-versus-gateway comparison in the cluster: Coinbase operates both the sequencer and the chain-official `mainnet.base.org`, so the house endpoint has every locational advantage, and it still has to beat PublicNode, dRPC, Tenderly and Nodies on a level probe. 5 providers, the same `eth_getBlockByNumber` call every 60 seconds, three regions, stale-head detection against the cross-provider tip.
Methodology
Base: 5 free public RPC endpoints answering `eth_getBlockByNumber` every 60 seconds from us-east, eu-west and Singapore, ranked by 24h median. PublicNode leads at 76 ms; the chain-official endpoint measures 102 ms. The same page carries the private cohort (API key), Alchemy, Chainstack, GetBlock, Quicknode and Dwellir on their keyed Base endpoints every 120 seconds: switch the Access selector; Chainstack leads it at 5 ms. We measure the round-trip latency of a single, identical RPC call (`eth_getBlockByNumber`) against every no-key public Base endpoint that sustains continuous probing, 5 providers, every 60 seconds, from us-east, eu-west and Singapore. The harness also classifies every response (ok / http_err / jsonrpc_err / stale / timeout) and audits archive depth every 5 minutes, so the leaderboard rewards sustained, honest availability rather than a fast error message. The cross-chain view lives on the parent rpc-capabilities benchmark; this page is the Base-scoped answer with per-region breakdowns as a first-class dimension.
- Cadence: every 60 seconds per provider, from each of 3 probe regions (us-east Virginia, eu-west Amsterdam, sgp Singapore). Headline p50/p90/p99 aggregate across all 3 regions via Prometheus `avg(quantile_over_time(...))`; per-region breakdowns are first-class on this page via the region tabs.
- Payload: `{"jsonrpc":"2.0","id":<rotating>,"method":"eth_getBlockByNumber","params":["latest",false]}`. Plain HTTP POST, identical for every endpoint, no API key in any request. The request id rotates, but that does not guarantee a cache miss: many caches normalise `id` away. Measured 2026-10-08, fixed and rotating ids answer within 2 to 4 percent of each other.
- Latency measurement, call-result classification (`ok`, `http_err`, `jsonrpc_err`, `stale`, `timeout`) and the archive-depth audit follow the cluster method documented on [rpc-capabilities](openchainbench.com/benchmarks/rpc-capabilities#methodology); this page is the Base-scoped view of the same harness, with the same exclusion rules.
- Chain scope: every query on this page is pinned to chain="base". Provider coverage: 5 no-key public-read endpoints live today, each listed with its URL in the endpoints table on this page. Blink (formerly Merkle) was delisted 2026-07: its gateway does not serve `eth_getBlockByNumber`, the anti-cache probe this cluster standardized on, so it cannot be measured on equal terms. Exclusions follow the cluster-wide rules documented on the parent benchmark.
- Private cohort (Access selector, Private): Alchemy, Chainstack, GetBlock, Quicknode, Dwellir and BlockMachine probed on their keyed Base endpoints every 120 seconds from the same 3 regions, same call (eth_getBlockByNumber, rotating id), same ok / http_err / jsonrpc_err / stale / timeout classification.
- Keys live in the probe environment, never in the repo. Plan tiers, disclosed: Alchemy: an Enterprise trial key, measured on that tier since 10 November 2026, not a reader's tier; Alchemy states free and paid keys share one routing; Chainstack: a Global Node on the Growth plan; Quicknode: an endpoint on a Build plan ($49 per month), shared fleet; GetBlock: an endpoint on the Enterprise plan, shared by the three probe regions. Chainstack, GetBlock and Quicknode provided their endpoints.
- Dwellir plan tier, disclosed: a Developer plan key ($49 per month, 25 million API responses, 100 responses per second; 1 RPC response = 1 API credit, no compute units). Shared nodes, the same product a reader buys. Dwellir publishes no Solana endpoint, so it is absent from that page rather than shown as unavailable.
- GetBlock regional pairing: each probe region uses GetBlock's nearest endpoint, us-east to New York, eu-west to Frankfurt, sgp to Singapore. Alchemy, Chainstack and Quicknode publish one routed entry point per chain instead, which in the general case does land a probe on a nearby node. Some of any gap between the two arrangements is deployment topology rather than node speed, and a deployment in one of these regions would see the same benefit GetBlock's endpoint gives the probe.
- Two cohorts, two rankings: the public endpoints are shared, rate-limited gateways probed every 60 seconds; the keyed endpoints are metered or dedicated and probed every 120 seconds. The selector switches the whole page (ranking, chart, region tabs, badges, citation) between them. No figure on this page compares a public endpoint with a keyed one.
Results: 5 free public Base RPC endpoints ranked by p50 latency (24h, 3 regions)
| â„– | Providers | p50 | p90 | p99 | Success |
|---|---|---|---|---|---|
| 1 | PublicNodeAllnodes-operated, 70+ chains, archive on most | 76 ms | 96 ms | 235 ms | 100.00% (n=4320) |
| 2 | dRPCDecentralized RPC mesh, consensus-checked | 96 ms | 107 ms | 166 ms | 96.11% (n=4320) |
| 3 | TenderlyMulti-chain public gateway, 9 chains, no key | 101 ms | 108 ms | 116 ms | 100.00% (n=4320) |
| 4 | Base (Coinbase)Coinbase-operated, Base mainnet RPC | 102 ms | 107 ms | 124 ms | 100.00% (n=4321) |
| 5 | NodiesPOKT Network's decentralized public RPC successor, 7+ chains | 135 ms | 152 ms | 181 ms | 99.98% (n=4319) |
Public Base RPC endpoints measured
Chain ID 8453, currency ETH. The 5 no-key endpoints answering our probes today, with their current 24h median. Paste one into a wallet or a client as is: no signup, no key. Alchemy, Chainstack, GetBlock, Quicknode, Dwellir and BlockMachine, private endpoints that need an API key, are ranked separately under the Private tab; keyed URLs are never listed.
PublicNodehttps://base-rpc.publicnode.com76 ms100.0%
dRPChttps://base.drpc.org96 ms96.1%Tenderly
https://gateway.tenderly.co/public/base101 ms100.0%
Base (Coinbase)https://mainnet.base.org102 ms100.0%
Nodieshttps://base-pokt.nodies.app135 ms100.0%
Median round-trip and success rate over the last 24 hours across the probe regions; the ranked table above carries p90, p99 and the per-region split.
Not listed on Base
- Ankr: public endpoint requires an API key
- NodeReal: requires an API key
- 1RPC: delisted for reliability: under 13 % success at one probe per minute (since 2026-07-09)
- gateway.fm: 29 requests per IP budget, too small for continuous probing
- LlamaRPC: Cloudflare 521 from the probe regions
- BlockPI: Cloudflare 521 from the probe regions
- OmniaTech: Cloudflare 521 from the probe regions
- Alchemy demo: rate-limited public demo endpoint; Alchemy's keyed endpoint is in the private cohort on this page
Frequently asked
Which Base RPCs work without an API key?
The 5 providers on this page are listed with their URLs in the endpoints table above. Blink (formerly Merkle) was delisted in July 2026 because its gateway does not serve `eth_getBlockByNumber` (JSON-RPC -32601), the anti-cache probe used across the cluster, so its latency cannot be compared on equal terms. Every (provider, chain) pair was live-verified no-key before inclusion, and anything that key-gates, region-blocks or rate-limits below our 60-second cadence is excluded rather than listed with an asterisk.
Is Coinbase's official Base RPC faster than third-party gateways?
Compare the two numbers: despite being operated by the same team that runs the sequencer, `mainnet.base.org` measures 102 ms against 76 ms for PublicNode on the cross-region median, and the ranking moves with the region the request originates from. Check the region tabs on this page for the origin closest to your deployment; the cross-region average hides these flips.
Which Base RPC provider is fastest with an API key?
Per the live data: Chainstack at 5 ms (eth_getBlockByNumber p50 over 24h, averaged across us-east, eu-west and Singapore); Alchemy measures 9 ms, Chainstack 5 ms and Quicknode 6 ms; GetBlock measures 9 ms. On the free cohort PublicNode leads at 76 ms, measured on the same call from the same three regions; the two cohorts are never ranked in one table (shared gateways at 60 s against metered endpoints at 120 s). Switch the Access selector to Private for the region tabs and the p99 column; plan tiers are disclosed in the methodology.
Source code github.com/ChainBench/OpenChainBench/tree/main/harnesses/rpc-capabilities