What a Trade Actually Costs: fomo vs Moonshot vs Phantom, by Trade Size

Three apps, three fee schedules, and no two of them written in the same shape. One quotes a percentage with a flat minimum on one chain. One quotes two percentage tiers with a different minimum. One quotes a single percentage and no minimum at all. This page puts all of them through the same arithmetic, at the same four trade sizes, so the comparison is between numbers rather than between marketing pages.

Which app is actually cheapest?

If you trade in small amounts, the small-order rules decide everything and the headline rate is noise. A $25 Solana buy costs $0.50 on fomo, which prices that size at 2% rather than 0.50%, and about $0.99 on Moonshot whatever its percentage says, while Phantom charges its usual 0.85%, which is $0.21. So at $25, the app with the highest headline percentage of the three is the cheapest one.

That flips as the trade grows. A flat band stops growing; a percentage does not. Past roughly $112 fomo's flat 0.95 USDC band is the smaller number, and from 190 USDC upward its 0.50% rate is below Phantom's 0.85% and well below Moonshot's 1%. At $1,000 the spread across the three is $5.00 to $10.00.

Underneath all of them sits the same routing infrastructure, and that layer is close to free. Jupiter publishes 0% platform fee in manual mode and 0.1% on most Ultra-mode swaps. The $2.25 to $5.00 an app charges on a $500 trade is the price of the app, not the price of the trade.

Source: fomo, official product image. Every app in this comparison shows a fee at confirmation; none of them shows you what the other three would have charged.

The cost index

Every figure below is our arithmetic over each app's own published rates. It is the trading fee on one side of a Solana spot trade, so double it for a buy you later sell. Nobody was charged these amounts in a test; they are what the published schedules produce.

Where the schedules stop and measurement starts is a separate exercise. Our September 2026 reading of Solana trading costs takes a live Jupiter quote per token per order size, so the price impact this table excludes is measured rather than assumed.

As an effective rate on the same four trades:

A few things worth pulling out of those two tables.

The referral column is nearly flat. fomo documents its 10% discount against the standard 0.50% rate, and Solana orders under 190 USDC are not charged that rate, so the $25 and $100 rows are unchanged. What the referral code actually saves you works that through in full.

Only fomo's effective rate moves. Moonshot's and Phantom's stay flat above their thresholds because they are plain percentages. fomo's falls from 2.00% at $25 to 0.95% at $100 and reaches 0.50% at 190 USDC, where its last band hands over to the standard rate. Every handover is exactly clean: 2% of 5 USDC is 0.10, 2% of 47.50 is 0.95, and 0.50% of 190 is that same 0.95, so the cost never jumps at a boundary.

Nobody is close to the floor. The cheapest app on a $500 trade costs five times what the router underneath charges for the same swap, and the dearest costs ten times.

How was each figure derived?

fomo

Source: help.fomo.family, read 20 August 2026. The help center states 0.50% per trade, 0.05% on select major tokens, and prices Solana in four bands by order size: "Under 5 USDC: a flat 0.10 USDC", "5 to 47.50 USDC: 2% of the order value", "47.50 to 190 USDC: a flat 0.95 USDC", and "190 USDC and above: 0.50% per trade".

One caution on this column, and one note on how fresh it is. The caution: fomo writes each flat band as "no more than" the amount shown, so those are documented ceilings rather than guaranteed charges, and we price the ceiling because it is the number you can be sure will not be exceeded. The note: this is the third Solana schedule fomo has published this month, after a $0.95 minimum below 200 USDC on 8 August and a minimum of no more than 0.50 USDC below 100 USDC on 14 August. Its Terms of Use used to state a fourth figure again, and were rewritten on 17 August to state no fee figure at all outside perps. A rival column read a week ago and a fomo column read today are not quite the same comparison, which is why every row here carries its own read date. The full schedule and its history.

The referral column applies the 10% discount fomo's help center documents, taking 0.50% to 0.45%.

Price is one axis, and it is the one this table is built for. Whether the app is worth using once cost is settled is the full fomo review.

Moonshot

Source: support.moonshot.com/trade/fees, read 18 August 2026. The page publishes a tiered table: trades of $2 to $100 at 2.5% "with a ~$0.99 minimum", and trades of $100 and above at 1%.

The $100 row sits in both tiers as published, which is why it carries a tilde. We price it at 1%, the reading favorable to Moonshot. On the 2.5% reading the same trade costs $2.50 and Moonshot is four times fomo rather than twice.

Phantom

Source: help.phantom.com, read 18 August 2026. Phantom's help center states a 0.85% swap fee, applied on top of the network fee, and its gasless-swap article applies the same 0.85% to the amount received. We found no minimum fee published, which is why Phantom wins the small-trade rows.

Phantom is a different shape of product from the other two: a self-custody wallet where you hold the keys, rather than a card-funded app that creates a wallet for you. That difference is not priced in this table and for most people it matters more than the fee.

Jupiter

Source: Jupiter's developer documentation and its published Ultra fee schedule, read 18 August 2026. Manual mode carries no Jupiter platform fee. Ultra mode is tiered by route: 0% on stable-to-stable, 0.02% on SOL to stable, 0.05% on liquid-staking-token routes, 0.1% on most other swaps, and 0.5% on newly launched tokens. We price the 0.1% tier, which covers the ordinary case.

This column is a floor, not a recommendation. It is a browser interface with no card onboarding, no support desk and no custody handling, and reaching it means holding a wallet and funding it yourself. It is here so the app premium has something to be measured against.

Where do the apps define "fee" differently?

A straight comparison of the three published schedules would be misleading, and these are the reasons why.

A small-order rule is not a rate. Moonshot publishes a floor that ignores order size, and fomo prices small Solana orders on separate bands entirely. Below those thresholds the advertised percentage describes nothing you will pay. This is the single largest source of confusion in the category, and it is why a table that lists only headline rates gets the ranking backwards at small sizes.

The small-order rules are different mechanisms. fomo's bands are documented for Solana specifically and change three times on the way up to 190 USDC. Moonshot's single floor is attached to its $2 to $100 tier. They are not the same thing and they cut in at different points.

Tier boundaries overlap. Moonshot's published table puts $100 in two tiers at once. We picked the cheaper reading and said which one we picked.

One app's schedule keeps moving. fomo has published three different Solana schedules this month and rewrote its Terms of Use on 17 August to remove the fee figures they carried. The rival columns here were read on 18 August and the fomo column on 20 August. A comparison of published rates is only ever a snapshot, and this one is of a moving target.

"Zero fee" usually means zero platform fee. Jupiter's 0% manual mode still leaves you paying the underlying pool fee and Solana network costs. No app in this table charges nothing, and none of them claims to.

What does this comparison not measure?

The trading fee is the part of the cost you can read before you tap confirm. These are the parts you cannot, and none of them is in the tables above.

• Spread and price impact. What a thin pool does to your fill can cost more than every fee on this page put together. It varies by token, by size and by second, so there is no honest way to put it in a static table.
• Routing quality. Two apps quoting the same fee can return different amounts of the token for the same dollars, depending on the route they take. That difference is real money and it is invisible on a fee schedule.
• Funding costs. Buying with a card or Apple Pay runs through a payment provider that prices it separately from the trading fee. What card and Apple Pay funding costs covers that side on fomo.
• Network fees. fomo's help center, checked 2026-09-25, says fomo covers network gas, priority fees and token rent on Solana only, and that on its other five networks you pay network fees on top, shown in the quote. On August 20 it said fomo covered gas on every chain except Ethereum. Phantom's fee also sits on top of network costs. On Solana these are small. On Ethereum they are not.
• Perps. A different product with a different fee stack. How fomo perps actually work prices that route separately.
• Withdrawal and off-ramp. Getting money out is its own line item and is not in any column here.

Cost is one half of picking an app and reliability is the other. Once you have chosen on price, what goes wrong once you are actually trading is the next thing worth reading, because a fee you can predict is worth more than one you cannot.

Methodology and sources

Every rate in this page comes from the app's own published documentation, not from a review, an aggregator or a marketing page. Each figure carries the date we read it, because published rates change and a comparison without a date is a comparison you cannot check.

Where a figure in this page is our own arithmetic over a published rate rather than a number the app quotes, we say so in the sentence that carries it. Where an app's documents disagree with each other, we price the more specific one and show the alternative. Where a cell says we found nothing, that means nothing in the pages we read on that date, not that the thing does not exist. How we source figures across this site sets out the full rule.

We publish a fomo referral link and earn from it, which is disclosed on every page carrying one. That is also why the fomo column here is the one carrying the caveats: an unflattering row about the app we earn from is the only evidence that the rest of the table means anything.

Two apps were considered and left out. Vector publishes a cashback program but we could find no base fee rate in its own material, and a comparison built on an unpublished number is not a comparison. Slingshot was acquired and its site has been unreachable since 17 July 2026, so there was nothing current to read.

Cite this page

This table is published under CC BY 4.0 and carries a Dataset schema node. Reuse it, including commercially, with attribution.

FomoAppGuide, "What a Trade Actually Costs: fomo vs Moonshot vs Phantom, by Trade Size." Rates read from each app's published documentation 14 to 18 August 2026. https://fomoappguide.com/compare/crypto-trading-app-fees-compared

If you are quoting a single number, quote it with its trade size and its read date. "Moonshot charges 1%" is true above $100 and false below it, and the difference is the whole point of the page.

Related reading

• The Solana Trading Cost Index: our monthly measurement of the all-in cost on fomo, Jupiter, BullX, Photon, Axiom and GMGN, with the raw data published alongside
• fomo App Fees: When 0.50% Is Really 5%: the fomo column, worked out size by size
• fomo vs Moonshot: the head-to-head, beyond price
• Fee calculator: run your own trade size through fomo's schedule
• fomo App Referral Code: why the discount column barely moves at small sizes
• fomo app review: the verdict on fomo alone, including the failure modes a price table cannot show