Katana RPC endpoints: free public URLs by latency
HTTP round-trip latency for the `eth_getBlockByNumber(\"latest\", false)` method against every free, no-key public Katana EVM endpoint, audited every 60 seconds from 3 regions.
TL;DR. As of , dRPC has the lowest median latency of the 3 free public Katana RPC endpoints measured, 113 ms (p50, 24h, 3 regions). Source: OpenChainBench, https://openchainbench.com/benchmarks/katana-rpc. 3 public endpoints, no key: katana.drpc.org, katana.gateway.tenderly.co, rpc.katana.network. Full URLs with their 24h median are listed below.
Companion page
Comparing free RPC endpoints across every chain we measure? Open the cross-chain RPC matrix →
Katana is an EVM L2 (chain 747474) aimed at DeFi, reachable through 3 keyless endpoints. The 2026-10-07 sweep found the widest usable roster of any chain added that day alongside 0G and TAC: the Katana official RPC (rpc.katana.network), a Tenderly gateway (katana.gateway.tenderly.co), dRPC (katana.drpc.org) and Thirdweb (747474.rpc.thirdweb.com). All four returned chain id 0xb67d2 on the audit. Carrying both a managed gateway and a routing pool is unusual at this size, and it makes the per-region comparison worth reading: a managed gateway and a routing pool answer the same call very differently depending on where the request starts.
Methodology
Katana: 3 free public RPC endpoints answering `eth_getBlockByNumber` every 60 seconds from us-east, eu-west and Singapore, ranked by 24h median. dRPC leads at 113 ms; the chain-official endpoint measures 133 ms. We measure the round-trip latency of a single, identical JSON-RPC call (`eth_getBlockByNumber(\"latest\", false)`) against every no-key public Katana endpoint that sustains continuous probing, 3 providers measured, every 60 seconds, from us-east, eu-west and Singapore. The harness classifies every response (ok / http_err / jsonrpc_err / stale / timeout) with a Katana-scaled staleness gap (20 blocks at Katana's ~1 s block time). The cross-chain view lives on the parent `rpc-capabilities` benchmark; this page is the Katana-scoped answer with per-region breakdowns as a first-class dimension.
Results: 3 free public Katana RPC endpoints ranked by p50 latency (24h, 3 regions)
Public Katana RPC endpoints measured
The 3 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. Private providers (API key) are never listed with a URL.
dRPChttps://katana.drpc.org113 ms100.0%
Tenderlyhttps://katana.gateway.tenderly.co123 ms100.0%
Katanahttps://rpc.katana.network133 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.
Declared, but not answering our probes
This endpoint is named in the spec and probed on the same schedule as the rest, and answers too rarely to carry a median. Below 5 % success we leave it out of the table rather than publish a latency drawn from a handful of replies. A provider can serve a browser and refuse a datacenter, so this says what a server deployment would meet, not what the endpoint is capable of.
- Thirdweb
https://747474.rpc.thirdweb.com0.3 % success over 24 h
Frequently asked
Which Katana RPCs work without an API key?
4 endpoints are probed on this chain: Katana (rpc.katana.network), Tenderly (katana.gateway.tenderly.co), dRPC (katana.drpc.org), Thirdweb (747474.rpc.thirdweb.com). Every listed endpoint was live-verified with a `eth_getBlockByNumber` call returning a parsable response before inclusion.
How is Katana RPC latency measured here, technically?
One identical `eth_getBlockByNumber` call every 60 seconds against each provider from each of 3 regions, using the same plain HTTP client the rest of the RPC cluster uses. Wall-clock round-trip is recorded at millisecond precision; p50/p90/p99 are computed via Prometheus `quantile_over_time` over 24 hours. Responses are classified (`ok` / `http_err` / `jsonrpc_err` / `stale` / `timeout`) so an endpoint stuck on an old head or returning errors behind HTTP 200 is never ranked as fastest.
Why does Katana have both a Tenderly gateway and dRPC?
They are different shapes of infrastructure answering the same call. Tenderly fronts its own nodes behind a managed gateway; dRPC routes to a pool of third-party nodes. Both are keyless on Katana, which is rare, and the region tabs are where the difference shows: a managed gateway tends to hold a tighter distribution while a routing pool can win on median from one origin and lose badly from another.
Source code github.com/ChainBench/OpenChainBench/tree/main/harnesses/rpc-capabilities