MEV on Solana: Sandwich Attacks, Jito Bundles, and How Traders Get Affected
Maximum Extractable Value on Solana explained. How sandwich attacks, backrunning, and Jito bundles work — and what regular traders can do about slippage and routing.
Every time you trade on Solana, a validator decides in milliseconds what order transactions execute in. That order has a price — and whoever controls it can pocket it.
In plain terms: Imagine you’re at a supermarket self-checkout and someone behind the counter can see your entire basket before you scan anything. They dart to the shelf, grab the last item you need, and sell it back to you slightly pricier. That’s front-running on a blockchain: a bot sees your buy order, jumps in front of you, pushes the price up, and you end up paying more than you should have. MEV — Maximum Extractable Value — is the umbrella term for every strategy where bots and validators extract profit by controlling transaction order. Some forms (arbitrage) keep markets efficient. Others (sandwich attacks) come directly out of regular traders’ pockets.
Core idea
MEV stands for Maximum Extractable Value (originally “Miner Extractable Value” on Ethereum). It describes the profit that block producers, validators, and specialized bots can extract by reordering, including, or excluding transactions inside a block. Whoever decides the order of transactions in a block can earn money — by buying just before a big trade and selling right after, by being first to liquidate a vulnerable position, or by capturing an arbitrage between two DEXes. MEV is not a bug. It is an emergent property of every public blockchain that allows pending transactions to be observed and reordered. The question is how much value gets extracted, who gets it, and how transparent the process is. On Solana, Jito has turned previously chaotic MEV into a structured auction — with clear winners (searchers, validators, stakers) and clear losers (traders who leave slippage settings too loose).
Key facts
- MEV arises wherever transaction order translates into money
- Sandwich attacks cost regular traders directly
- Arbitrage and liquidations are necessary for functional markets
- ~80% of Solana validators (weighted by stake) run the Jito client — Jito is effectively Solana’s MEV layer
- JitoSOL stakers benefit indirectly from MEV tips; regular traders carry the risk
Not financial advice.
How MEV Works on Solana
Solana’s MEV landscape differs significantly from Ethereum’s:
No public mempool. Ethereum has a public mempool (a waiting room of unconfirmed transactions visible to anyone) — anyone can see pending transactions before they’re included. On Solana, transactions are forwarded directly to the current leader (the validator currently producing the next block). There’s no global pre-confirmation pool that searchers can scan.
Leader-based block production. Solana validators take turns producing blocks (4 consecutive slots per leader). MEV opportunities exist within the slot the leader is producing, and the leader chooses which transactions to include and in what order.
Jito’s role. Jito Labs operates a modified validator client (Jito-Solana, used by the majority of mainnet stake) that adds an off-chain auction layer for transaction bundles (atomic groups of transactions that execute together or not at all). Searchers submit bundles with tips; the leader includes the highest-bid bundle. This turns Solana’s previously chaotic MEV into a structured auction.
Speed matters. Solana’s sub-400ms block times mean MEV strategies must be executed by colocated bots with low-latency RPC and direct validator connections. There is no time for slow searchers.
MEV Types on Solana
Sandwich Attacks
A bot detects a large pending swap, places a buy in front of it (pushing the price up), lets the victim’s swap execute at the worse price, then sells immediately after. The bot pockets the spread; the victim eats the slippage.
On Solana, sandwich attacks require:
- Visibility into the victim’s transaction (via private RPC relationships or direct validator access)
- A bundle that frontruns AND backruns the victim atomically
- Jito tip high enough to win the auction
Sandwich attacks are most common on memecoin trades through Pump.fun and Raydium pools where slippage settings are loose.
Backrunning
Less aggressive than sandwiching: the bot just places a buy or arbitrage trade immediately after a price-moving transaction. No frontrun, no direct victim cost — but the bot captures the arbitrage opportunity that the original trade created.
Backrunning is the most common form of MEV on Solana and is generally considered a “neutral” extraction (no one is directly worse off, but the arb is no longer available to anyone else).
Liquidations
When a borrowing position on Kamino, MarginFi, or Solend becomes undercollateralized, anyone can trigger the liquidation and earn a bonus (typically 5-10% of the liquidated collateral). Specialized liquidator bots monitor on-chain state continuously; whoever sends the liquidation transaction first wins.
Arbitrage
Cross-DEX arbitrage: token X trades at one price on Raydium and a different price on Orca. A bot buys on the cheaper venue, sells on the more expensive one, captures the spread. Arbitrage is generally considered “good” MEV — it keeps prices aligned across markets.
See the Solana DEX Comparison for a breakdown of the major venues bots arbitrage between.
Jito Bundles and MEV Auctions
Jito introduced the concept of bundles to Solana — atomic groups of transactions that either execute fully in order or not at all. Bundles include a tip transaction; searchers compete for inclusion by raising tips.
How a Jito bundle works:
- Searcher constructs a bundle (e.g., frontrun + victim’s swap + backrun, or arb step 1 + arb step 2)
- Bundle is submitted to a Jito Block Engine
- Multiple searchers’ bundles compete in an auction; highest tip wins
- Winning bundle is forwarded to the current leader for inclusion
Why this matters:
- Atomicity: the bundle either executes fully or reverts — no partial fills that leave the searcher exposed
- Tips funnel back to validators and stakers: see distribution section below
- Transparency: Jito publishes auction data at jito.wtf, making MEV flows on Solana more visible than they are on most chains
Jito’s auction model has effectively replaced the previous “spam-the-RPC” race with a structured market. Bundles can be inspected after the fact via Jito’s explorer.
How to Protect Yourself as a Regular Trader
You cannot eliminate MEV exposure, but you can reduce it:
1. Set tight slippage on swaps. Slippage tolerance (the maximum price deviation you accept on a swap) is the attack surface. If your slippage tolerance is 1%, a sandwich bot has 1% of room to extract value from you. If it’s 0.5%, the attack window shrinks. Many DEX aggregators default to 0.5-1%; for memecoins with low liquidity, traders often set 5-15% — that’s a generous window for sandwiches.
2. Use MEV-aware aggregators. Jupiter is the dominant Solana DEX aggregator. It routes across multiple venues and offers a “swap mode” with dynamic slippage protection. Jupiter does not eliminate MEV but makes individual trades harder to target predictably.
3. Use private RPC submission for large trades. Some RPC providers (including Helius) offer transaction submission paths that don’t expose your transaction to public observers before block inclusion. For trades large enough to be profitable to sandwich, this matters.
4. Avoid extreme price impact. A swap that moves the pool price by 5%+ is highly attractive to sandwich bots. Splitting into smaller orders or using time-weighted strategies reduces the signal.
5. Recognize when you’re being targeted. If your trades on a specific token consistently fill at the worst end of your slippage range, you may be a known target. Switch tools, use a fresh wallet path, or change venues. See Detecting Solana Bots for adjacent forensics patterns.
MEV Distribution — Who Profits?
MEV value flows through three layers:
Searchers. Independent bot operators who identify opportunities (sandwiches, arbitrage, liquidations) and submit bundles to Jito. Searchers pay tips; their profit is the MEV extracted minus the tip.
Validators. Jito-Solana validators receive the bundle tips. Tips are added to block rewards and distributed proportionally to the validator’s stakers (after a validator commission).
Stakers via Jito. Holders of JitoSOL — the liquid-staking token issued by Jito — earn a share of MEV tips on top of standard staking rewards. This was Jito’s mechanism for redistributing MEV revenue back to ordinary stakers rather than concentrating it with sophisticated searchers and validators alone.
According to Jito’s public dashboards, MEV tips have historically added a meaningful APY uplift on JitoSOL versus non-MEV-aware liquid-staking tokens. Exact figures fluctuate with on-chain activity — see jito.wtf for current data.
For a broader view of liquid-staking on Solana, see Liquid Staking on Solana.
FAQ
Is MEV illegal?
No. MEV is the natural result of how blockchains order transactions. Sandwich attacks are aggressive but not against any protocol rule. Some jurisdictions are studying whether predatory MEV constitutes market manipulation, but as of 2026 there is no settled regulatory position.
Does Solana have less MEV than Ethereum?
Different, not necessarily less. Solana’s no-public-mempool design eliminates one entire class of MEV (mempool sniping by anyone) but introduces a different class (bundle auctions controlled by Jito-aligned infrastructure). Total MEV value extracted on Solana is non-trivial.
Can I run a searcher bot myself?
Technically yes. Practically: searchers compete on latency, capital, and signal quality. The barrier to profitability is high. Most public information about searcher economics is on the Helius blog and Jito’s docs.
What’s the difference between MEV and gas auctions?
Gas auctions are about getting included in a block. MEV auctions are about what order transactions appear in once included. Solana’s fee market plus Jito tips effectively combines both into one auction.
What this means for you
MEV is not a niche edge case — it is a structural feature of every public blockchain. On Solana, sandwich attacks hit traders with large orders and loose slippage settings hardest. Arbitrage bots, by contrast, keep prices aligned across markets and make execution fairer on average. Holders of JitoSOL participate passively in MEV revenue. The whole system is transparent and publicly auditable via jito.wtf — which puts Solana ahead of most chains in MEV visibility.
Want to put this into practice safely? This article covers the concept. The step-by-step path — wallet setup, security, staking, DeFi, and taxes — is in the Solana Guide.
Sources and Further Reading
Jito and bundles
- Jito: jito.network — Jito-Solana validator client and bundle infrastructure
- Jito explorer: explorer.jito.wtf — public bundle and tip data
- Jito stats dashboard: jito.wtf — MEV revenue and validator metrics
- JitoSOL: jito.network/jitosol — MEV-aware liquid staking
Tech background
- Helius blog on MEV: helius.dev/blog
- Helius transaction sender: helius.dev — RPC paths for Solana
- Jupiter: jup.ag — DEX aggregator with dynamic slippage
Academic and research sources
- Daian et al. — Flash Boys 2.0 (2019) — foundational MEV paper
- Babel et al. — Clockwork Finance (2021) — MEV formal analysis
- Flashbots Research — ongoing MEV research (Ethereum-focused but conceptually relevant)
Next Steps
- DEX activity and routing: Solana DEX Comparison
- Wallet-level forensics: Detecting Solana Bots
- MEV-aware staking: Liquid Staking on Solana