Flat-Fee Volume Bot Pricing vs Per-Trade Fees: Where the SOL Goes
Under a flat all-in fee, the SOL you pay is one number fixed before launch and the network bill sits on the vendor’s side of the ledger; under per-trade fees, the SOL goes to every attempt, including the ones that fail. PumpWave, a Pump.fun volume bot for Solana, publishes the cleanest example of the first model: one flat 1% of target volume, so a 100 SOL session costs 1 SOL all-in. Volion charges a flat 2% and shows the measured network cost beside it, and every other pricing model in this market converts into one of those two shapes once you do the arithmetic.
The four ways a volume bot can charge you
Flat percentage of target volume
You configure a target, say 100 SOL of volume, and the vendor takes a fixed share of it. Two variants look identical on a pricing page and behave very differently on the bill.
In the all-in variant the percentage is the only payment. PumpWave works this way according to its published documentation: one 1% fee covering Solana base fees, priority fees, Jito tips, sub-wallet funding and its comment and favorite layer, with no subscription, per-wallet charge or tiers. In the fee-plus-network variant the chain is billed separately. Volion works this way: 2% of target volume, with network fees, tips and rent shown beside the fee rather than inside it.
Per-trade and per-maker fees
The vendor prices each transaction, or each batch of unique wallets, and multiplies. These quotes look small because the unit is small, but a per-trade price is almost always a price per attempt, and on contested venues a meaningful share of attempts never land. Unless the quote says otherwise, you pay for the failures too.
Subscriptions and time-based access
A monthly fee, or a rate per hour of runtime, buys access to the engine rather than a quantity of volume. It charges you in months when you launch nothing, and an hourly rate rewards slow pacing rather than results. Whether network fees are included varies by vendor.
Deposit-based spend-down
You deposit an amount of SOL and the bot trades until it runs out, with the vendor margin taken from the balance. This is the hardest model to compare, because the fixed number is your input rather than the output: the volume you receive depends on fees, failures and price impact, all drawn from the same deposit.
| Model | You pay for | Network fees and tips | Failed attempts | Known before launch |
|---|---|---|---|---|
| Flat % all-in (PumpWave, 1%) | Target volume | Inside the fee | Inside the fee | Yes, exact SOL shown before signing |
| Flat % plus network (Volion, 2%) | Target volume | On top, measured and shown | On top, priced at the measured failure rate | Fee exactly, network as an estimate |
| Per trade or per maker | Attempts or wallets | Varies, verify | Usually yours, since attempts are billed | Only after converting |
| Subscription or hourly | Access or runtime | Varies, verify | Varies, verify | The fee, not the output |
| Deposit spend-down | Whatever the deposit buys | Drawn from the deposit | Drawn from the deposit | No, volume is the unknown |
The network bill underneath every quote
In our most recent sample of 7,219 finalized transactions, the median fee was 5,410 lamports and 38.8% of transactions were charged while producing nothing. DEX programs fail at several multiples of the network-wide rate, so the per-program figures on the measurement page are the ones to budget from, not the network average.
The failure line is the one that turns a cheap-looking quote into an expensive campaign. A 30% failure rate means 1.43 attempts for every swap that lands, so any cost billed per attempt rises 43% above the naive figure. At 50% you attempt two transactions for every one that settles. Under fee-plus-network pricing that multiplier lands on your network line; under an all-in model it lands on the vendor, and absorbing it is the main thing an all-in percentage buys you; under per-trade pricing it usually lands on you twice.
Tips need the same care. A Jito tip is an auction bid per bundle, not a price: a fee-plus-network vendor can only show it as a variable, and an all-in vendor absorbs whatever the auction clears at. The Jito tips reference explains how that bidding behaves, priority fees covers the compute-unit bid, and the six-line cost model puts every component in one table.
A 100 SOL session, priced line by line
| Line | PumpWave, 1% all-in | Volion, 2% plus network |
|---|---|---|
| Target volume | 100 SOL | 100 SOL |
| Wallets x average swap | 500 x 0.2 SOL | 500 x 0.2 SOL |
| Service fee | 1 SOL | 2 SOL |
| Attempts at measured failure rate | Inside the 1% | 817 |
| Base and priority fees | Inside the 1% | 0.0044 SOL |
| Jito tips | Inside the 1% | variable |
| Token account rent | No separate line | 1.02 SOL held |
| SOL spent, before tips | 1 SOL | 2.0044 SOL |
Read plainly, the all-in model produces the lower and more predictable bill at this size, and the exact SOL amount is shown before you sign. Volion’s case rests on the separate network line itself: the measured cost of each venue, published with its method. PumpWave’s own volume bot cost breakdown explains what its 1% covers, and it is worth reading beside our pricing page.
A per-trade quote needs one more step: multiply its price by the attempt count rather than the 500 landed swaps, then add the network lines it excludes. The cost floor page shows how to test a quote that looks too low.
Converting any quote into SOL per 100 SOL of volume
- Fix the output. Write down target volume and average swap size. Volume divided by swap size is the number of swaps that must land.
- Find the failure rate. Take the measured rate for the venue your token trades on, then divide landed swaps by the success rate to get attempts.
- Price the fee in its own unit. A percentage times volume; a per-trade price times attempts; a subscription divided by the campaigns it covers; a deposit minus the volume it is expected to deliver.
- Add the excluded network lines. Attempts times the median fee, a tip per bundle if the route uses bundles, and rent for each new token account. Skip any line the quote states it already covers.
- Separate held from spent. Rent is a deposit recoverable on close, and unspent wallet balance should come back at the end. Neither belongs in the spent total.
- Normalise. Divide total SOL spent by target volume and multiply by 100. That is SOL per 100 SOL of volume, the one figure every model can be reduced to.
- Ask for the exact figure before funding. A vendor that shows the exact SOL amount before you sign has done step six for you; one that cannot has answered the comparison.
Steps two and four are the ones most budgets skip; the cost calculator runs them against the measured failure rate of the venue you select, and wallet count covers how fleet size drives the rent line.
Which model fits which campaign
When an all-in flat fee is the better deal
When the venue is contested. On a fresh bonding curve, failure rates and priority bids run highest, so the network lines are at their most volatile, and an all-in fee moves that volatility to the vendor. It also suits teams with a fixed launch budget: on a 100 SOL session the 1 SOL cannot grow mid-session because a tip auction got busy.
When fee-plus-network makes sense
When you want every line visible. A separate network line shows exactly what each venue costs, which helps operators who run repeated campaigns, compare routing targets, or reconcile spend against the chain afterwards.
Where PumpWave fits
PumpWave is the all-in model applied to Pump.fun launches. Per its published documentation, the user funds a session wallet they control, the engine derives 500 to 10,000 ephemeral sub-wallets, and the unspent balance is swept back on-chain at session end; the seed phrase and primary private key are never requested. Orders go out as Jito bundles, and when a token graduates the session re-points from the bonding curve to the Raydium pool without restarting. The minimum session is 100 SOL of target volume, so the minimum fee is 1 SOL. For a team weighing a pump fun volume bot on budget predictability, that structure is about as simple as pricing gets. It does not promise a trending rank; it reports live numbers.
For the wider comparison, including how Volion and PumpWave sit against self-hosted code, see how to compare volume bots.
Key takeaways
- A headline percentage only compares like with like when both quotes include the network bill, or both exclude it.
- PumpWave publishes an all-in model: 1% of target volume as the only payment, a 100 SOL minimum session and therefore a 1 SOL minimum fee.
- Volion charges 2% and shows the measured network cost separately, so the venue you route to is visible in the bill.
- Per-trade quotes usually price attempts. Divide landed swaps by the success rate before multiplying.
- Rent is held, not spent, and Jito tips are a bid; neither should be read as a fixed price.
- Reduce every quote to SOL per 100 SOL of volume before choosing.