OpenChainBench
All answers
AnswerBacked by Fastest MEV protection RPC, live wallet latency and method coverage

Which MEV protection RPC is the most consistent?

Consistency means the p99 to p50 gap, not the median. On the live OpenChainBench MEV protection benchmark, Blink (formerly Merkle) holds the flattest tail of the cohort: its p99 sits within roughly 10 percent of its p50, while the fastest median gateway spikes an order of magnitude above its own p50 at the tail. For a wallet that fires a balance refresh, a gas estimate and a simulation on every screen, the tail is what users feel.

Data as of , refreshed continuously.

Wallet teams picking a default MEV protection RPC usually compare medians, because medians are what marketing pages publish. But a wallet calls its RPC constantly: every open screen triggers balance refreshes, gas estimates and eth_call simulations. At that call volume the user experiences the p99, not the p50. A gateway that answers in 33 ms most of the time but stalls to 400 ms on one call in a hundred produces a wallet that feels randomly broken several times per session. This page reads the live OpenChainBench mev-protect-rpc benchmark through the consistency lens: which of the public keyless gateways (Flashbots Protect, MEV Blocker, Blink, bloXroute Protect, BlockSec, 48 Club, PancakeSwap MEV Guard) keeps its tail closest to its median.

Live leaderboard, top 5

  1. bloXroute Protect logo

    bloXroute Protect

    #1 · Wallet call latency

    9msp99 13 ms
  2. 48 Club Privacy RPC logo

    48 Club Privacy RPC

    #2 · Wallet call latency

    22msp99 24 ms
  3. PancakeSwap MEV Guard logo

    PancakeSwap MEV Guard

    #3 · Wallet call latency

    26msp99 29 ms
  4. MEV Blocker logo

    MEV Blocker

    #4 · Wallet call latency

    33msp99 262 ms
  5. Blink logo

    Blink

    #5 · Wallet call latency

    97msp99 103 ms

Full live data: /benchmarks/mev-protect-rpc, refreshed every minute.

Methodology and data sources

Same data as the parent benchmark. Every 60 seconds, from us-east, eu-west and Singapore, the harness sends the 7-method wallet set (chainId, blockNumber, gasPrice, getBalance, call, estimateGas, feeHistory) to each public MEV protection gateway, keyless, one request per 1.5 seconds. The per-tick median across served methods feeds a Prometheus gauge; p50, p90 and p99 are quantile_over_time over 24 hours, averaged across regions. Consistency is read as the ratio between the p99 and p50 columns of the same provider. No transactions are sent; this is the read path a wallet exercises all day.

What this number does not tell you

  • ·Consistency of the read gateway says nothing about inclusion rates or refund economics, which are write-path properties requiring funded transactions to measure. OpenChainBench does not extrapolate them.
  • ·A flat tail with a moderate median can coexist with a fast median and a heavy tail. Which profile is better depends on the product: trading bots care about the median, consumer wallets about the tail.
  • ·Success rate must be read next to both numbers. A gateway can post flat latency while throttling a share of calls; throttled calls do not enter the latency distribution.

Frequently asked questions

Which MEV protection RPC has the flattest latency tail?
Blink (formerly Merkle) currently posts the flattest p99 to p50 ratio of the measured cohort on Ethereum, with a tail within roughly 10 percent of its median. The live columns on the benchmark page update every minute and the ranking can move; the ratio between the p99 and p50 columns is the number to read.
Why does consistency matter more than median latency for a wallet?
Because a wallet fires dozens of RPC calls per session. At 50 calls per session, a 1-in-100 stall hits most sessions at least once. The user does not experience the median; they experience the worst call on the screen they are currently looking at.
Is the fastest median MEV protect RPC also the most consistent?
Not currently. The gateway with the best p50 on Ethereum shows a p99 an order of magnitude above its own median, while the most consistent gateway trades a slower median for a nearly flat tail. The benchmark page shows both columns side by side so the trade-off is explicit.
Does this ranking cover Base and BSC too?
The parent benchmark measures Ethereum, Base and BSC as separate chain tabs. On Base only one MEV protection gateway is measurable keyless today (Blink); on BSC the cohort includes Blink, bloXroute Protect, 48 Club, PancakeSwap MEV Guard and BlockSec, so the consistency comparison is meaningful there as well.

Related questions

Same data as /benchmarks/mev-protect-rpc, refreshed every minute. Open methodology, open source.