Axiom MEV Protection: Off vs Reduced vs Secure Explained

MEV protection is the single setting on Axiom most likely to be left on a default that does not suit the trade, and the reporting on what that has cost Solana traders as a group is worse than most people realize.

Source: Axiom, official product image

What a Sandwich Attack Actually Is

A sandwich attack works by placing one transaction immediately before yours and one immediately after. The first pushes the price up right before your buy executes, so you pay more than you should; the second sells into the price impact your own buy created, capturing the difference. The structure depends on your order moving the price in a direction the attacker can profit from, which is why the classic version targets buys. That is a description of the mechanism, not a promise that a sell can never be exploited by some other technique.

The Three Modes

Off. No protection. Your transaction goes through the standard path with no defense against front-running. Axiom presents this as the fastest option and the most exposed.

Reduced. Routes your transaction through Jito's bundle system. Jito is a separate company and that bundle infrastructure is Jito's, not Axiom's; Axiom's Reduced mode is a route into it. Axiom describes this as adding protection while staying relatively fast.

Secure. Routes exclusively through whitelisted validators. Axiom's documentation describes this as the slowest of the three, trading away access to the fastest general-purpose path for a validator set selected to resist sandwich construction, and recommends it.

The Real Cost of Getting This Wrong

Cryptopolitan reported on 22 October 2025, citing on-chain research credited to 0xGhostLogs, that wide sandwich attacks extracted roughly 529,000 SOL across an estimated 222,272 victims over a year, with a peak of 87,000 SOL in January 2025, and that around 70% of the extracted SOL came from users on Solana's Axiom platform.

Two things about that figure are worth stating plainly. First, it is 70% of the SOL extracted, not 70% of the victims; the two are often conflated and they are not the same measurement. Second, the report does not establish a cause. Axiom's share of Solana terminal volume is a plausible part of it and traders leaving protection off is another, but neither the reporting nor this site can attribute the split, and none of it demonstrates that Axiom's infrastructure is uniquely vulnerable.

We link the reporting rather than the underlying analysis because we have not found a stable public URL for 0xGhostLogs' raw dataset. Where this figure appears elsewhere on this site, it carries the same attribution and the same caveat.

The same Cryptopolitan piece reports that the JitoSOL Blacklist Committee banned 15 further validators from receiving JitoSOL stake in response to that research. That is an action by Jito's committee on Jito's own staking product, and your own mode setting remains the more direct lever available to you.

When to Use Which Mode

This setting sits alongside your priority fee and Jito tip, both of which default to 0.001 SOL per Axiom's documentation, read August 2026, as the levers that determine execution quality on Axiom. See our full fee breakdown for how those interact, and how Pulse and Axiom's trading tools work for the rest of the execution layer.