Solana is getting ready to change a rule that has let its worst freeloaders off the hook for years — and if the plan passes, spam bots will finally pay for the mess they make, while everyday users could see transfers get about 20% cheaper. The proposal, known as SIMD-0553, cleared its first community vote on August 4 and is now in a roughly two-week discussion window that could reshape how the network charges everyone who uses it.
By Diego Rivera | August 14, 2026
The Hook: Why Solana’s Cheap Fees Have a Dark Side
Solana is famous for being fast and cheap — like an express lane on a highway where almost nobody pays a toll. But that bargain has a hidden cost. Right now, a transaction that does almost nothing costs the same as a transaction that hogs enormous amounts of the network’s computing power. The researcher behind the new proposal, known as Cavey, who works at Solana infrastructure firm Temporal, put it simply: if he submits a transaction that does nothing versus one that burns 200 million computing cycles, he is charged the same amount.
That flat-rate bargain is great for humans sending stablecoins. It is even better for robots. Trading bots fire off millions of transactions that mostly fail, pay pennies for the attempt, and leave everyone else to deal with the congestion — like shoppers filling every checkout lane with carts they abandon at the register.
The Evidence: The Numbers Behind the Freeloading
The data behind the proposal shows just how lopsided the current system is. According to Cavey, over the past 30 days the five traders with the highest failure rates submitted a staggering 11.5 million transactions. Those attempts consumed 929 million compute units across just 2,477 successful trades that generated about 16,091 dollars in profit — and for all of that, the bots paid only 78 SOL in fees. At current prices, that is a rounding error for the amount of network capacity they devoured.
- Simple transfers get cheaper — stablecoin and token transfers could become roughly 20% cheaper under the new model.
- Behind-the-scenes costs drop too — Temporal’s modeling suggests vote transactions would cost about 12.3% less and oracle data updates about 16.9% less.
- Heavy users pay the bill — a high-priority swap routed through DFlow would cost about 9.72% more, a mid-priority OKX swap about 301% more, and a zero-priority pump.fun swap an eye-watering 3150% more.
- Even the worst case stays cheap — Cavey argues the most compute-intensive transactions would still cost around 5 cents, compared with the 2 to 5 dollars a user might pay to swap 100 dollars on a centralized exchange.
The Core Conflict: More Burn or Bigger Validator Paychecks?
Here is where it gets contentious. Under the proposal, the new resource fee would not go to validators as income — it would be burned, meaning the SOL is permanently removed from circulation, like a bank shredding cash. Validators, the operators who run the network’s computers, could initially see base-fee revenue fall by around 4%. Not everyone is happy about that. One contributor, known as bji, argued on GitHub that more burn should not be a goal in itself and that validator income should not be arbitrarily reduced.
The burn angle is what makes this story interesting for anyone holding SOL. Solana currently burns about 648 SOL per day in transaction fees. At the proposed terminal fee rate, that could rise to roughly 7,500 to 9,000 SOL per day — a 12 to 14-fold increase, assuming resource demand stays the same. The network still issues about 60,000 new SOL per day, so this alone would not make Solana deflationary. But Cavey sees a path: if Solana wins, he says, there is a chance it could actually become a deflationary currency.
Market Implications: What This Means for Your Portfolio
Solana’s token trades around 75 dollars, well below the highs of past cycles. If you own SOL or use Solana apps, this proposal matters in two ways. First, lower costs for basic transfers could make the network friendlier for payments and everyday use — the use case crypto keeps promising. Second, a structurally higher burn ties the token’s value more directly to how much the network is actually used: more activity means more SOL destroyed, which tightens supply over time.
There are risks. If trading bots find Solana too expensive, some activity could migrate elsewhere, and validators squeezed by lower fee income may push back during the vote. The proposal still needs to clear its discussion phase, which typically lasts about seven epochs — roughly two weeks — before moving forward.
The Verdict
SIMD-0553 is a classic “make the polluter pay” reform dressed in blockchain clothing. Bots that turned Solana’s cheap blockspace into a free shooting range would finally face a bill, regular users would catch a discount, and long-term holders get a mechanism that shrinks supply as usage grows. It is not law yet — Solana’s new on-chain governance process will decide — but the direction is clear: the free lunch for spam is ending, and that is hard to argue against.
The cryptocurrency market remains highly volatile. This article is for informational purposes only and does not constitute financial advice.
Disclaimer: This article is for informational purposes only and does not constitute financial advice.
11.5 million failed transactions from just 5 bots in 30 days and they paid 78 SOL total for it. the freeloader era might actually end
929 million compute units torched for 16k in profit. and people still call solana congestion organic demand lmao
16k profit on 78 SOL of fees was the whole business model. this proposal just makes the denominator honest
Cheaper simple transfers sounds nice until you ask where that resource fee goes. Validators are not going to vote for less income.
That is the part that usually kills these proposals. SIMD-0553 already cleared its first vote though, so maybe the math works for them.
cleared its first vote and everyone relaxes. simd 228 also sailed through early votes before getting shelved in the fee market fight, first votes mean nothing
228 died in the fee market fight because it hit validator economics. 0553 is base fee structure, different political map. spammers paying proportional to compute burned is hard to lobby against
228 died because validators ate the cost. 0553 makes spammers pay, so validators vote yes without touching their own revenue. different math
validators voting yes on someone elses bill is the oldest playbook in governance. wait till the mev lobby notices the priority lane is next
fail_tx_farmer simd 228 comparison is fair but 0553 hits base fees not priority fees. different set of losers, maybe an easier vote
Depends on what the new fee actually charges. If five bots paid 78 SOL total for 11.5 million failed attempts, honest users have been subsidizing that spam for months.
five bots burning 11.5 million failed transactions is basically a denial of service with extra steps. making spam pay its own bill is basic network hygiene
11.5 million failed transactions from five bots is an insane number. price the compute they burn and that DoS becomes unprofitable overnight
20 percent cheaper simple transfers while the five worst bots finally get a bill matching 11.5 million failed attempts. the express lane stays cheap only if the toll booth actually goes up this time
20 percent cheaper simple transfers while bots finally pay for their spam is the first solana fee redesign that actually helps humans
cheaper base fees just means the spam moves to wherever its still free. watch the bot operators recalculate, not disappear
they wont disappear, theyll optimize to the new breakeven. even paying compute costs, MEV extraction on 11.5M attempts still prints if the hit rate is above zero
botwatch_er optimizing to the new breakeven is still a win. spam that pays its own compute is a tax, spam that doesnt is a subsidy, huge difference
sure, but mev lives in priority fees and 0553 hits compute pricing. the spam denominator gets honest while the mev game keeps its own lane
priority_pile keeping the mev lane untouched is why 0553 survives where 228 died. the political map matters more than the engineering in these votes
if base fees drop 20 percent and priority fees quietly eat the difference, simple transfers gain nothing. watching the mev lane is the right tell here
the two week discussion window is where this dies or survives. validators voting against their own spam revenue takes actual courage
fee_flatline courage is exactly the word. validators eating a spam revenue cut to save the network long term, precedent says they hesitate
78 SOL for 11.5 million failed attempts works out to fractions of a cent per spam tx. no wonder they kept farming, the penalty was a rounding error
78 SOL for 11.5 million failed attempts is about 0.0000068 per spam tx. no fee redesign survives contact with a number that small unless its brutal