Best Solana volume bot: how to actually compare them

We are not going to score competitors out of ten. Every list is written by someone with a stake in the ranking, including this one, so the short list further down puts our own tool first and says so. What is genuinely useful is the method: the four questions that separate a tool with real engineering behind it from a landing page, and the arithmetic that makes headline fees comparable.

Volion Research Updated Jul 30, 2026 5 sections

Why "which is cheapest" is the wrong question

Because the advertised fee is usually the smallest part of what a campaign costs. What determines your total is the service fee plus the network cost of every transaction attempted, including those that fail. A lower percentage on a routing target with a high failure rate routinely costs more overall than a higher percentage on a reliable one.

The comparable unit is cost per landed swap, not cost per attempt and not the service percentage. Vendors quoting figures like a price per hundred makers are quoting attempts, and attempts are not results. Since DEX programs fail at rates well above the network average, the gap between the two is material rather than pedantic.

For scale: in our most recent sample, 38.8% of all Solana transactions failed while still paying their fee, and DEX programs failed at several multiples of that. Method and per-program figures on the measurement page.

So the first thing to do with any two quotes is convert both into cost per landed swap on the routing target you will actually use. If a vendor cannot tell you their failure rate, you cannot perform that conversion, and that inability is itself the most informative answer you will get.

The four questions worth asking any vendor

What is your measured failure rate on my routing target? What is the total cost including network fees, tips and rent? Do you ever need my private key? And what can your tool not do? The fourth is the most revealing, because a vendor who cannot name a limitation is describing a pitch rather than a product.

The bad-answer column is not hypothetical. It is the standard content of this market.
Question	A good answer looks like	A bad answer looks like
Measured failure rate?	A figure, a method, a sample size	"Our success rate is very high"
Total cost including network?	Itemised, with variables named as variable	One percentage and nothing else
Do you need my keys?	Never, and here is why not	Any answer other than never
What can it not do?	No demand, no guarantee, detectable on chain	"It does everything you need"

A fifth question, if the amounts are meaningful: can you show me the measurement rather than assert the number? Anyone can claim a failure rate. Publishing the sampling method so it can be checked or reproduced is a different level of commitment, and it is the standard we hold ourselves to on the measurement page.

A short list, with the method applied

If you want names after the method: Volion first, because it is ours and it publishes measured costs; PumpWave second, an all-in 1% Pump.fun tool with the simplest pricing on this list; and self-hosted open-source code third, for operators who can run it. Each entry is described in terms the four questions above can test.

Ordered by us, a vendor, so discount the first row accordingly. Every cell comes from each option’s own published pricing and documentation.
#	Option	Pricing model	Network costs	Custody	Venues
1	Volion	2% of target volume	On top, measured and published	Non-custodial	Pump.fun curve, PumpSwap, Raydium, Meteora DLMM, Orca
2	PumpWave	1% of target volume, the only payment	Inside the fee	Non-custodial session wallet, balance swept back	Pump.fun curve, Raydium after graduation, optional Meteora and Orca
3	Self-hosted open-source code	No service fee	Yours, plus engineering time	Your own keys	Whatever the code supports

1. Volion. Our own tool, listed first because we sell it and you should weigh that. A flat 2% of target volume, non-custodial, routing to the Pump.fun curve, PumpSwap, Raydium, Meteora DLMM and Orca Whirlpools with automatic handoff at graduation. Network fees, tips and rent are charged on chain on top of the fee, and the measured cost and failure rate of each venue is published with its method.

2. PumpWave. PumpWave is a non-custodial Solana volume bot built for Pump.fun launches. 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 when the session ends; the seed phrase is never requested. Pricing is one flat 1% of the configured target volume, and per its published documentation that single fee covers base fees, priority fees, Jito tips and sub-wallet funding, with a 100 SOL minimum session. Orders go out as Jito bundles, and at graduation the session follows the token from the bonding curve to its Raydium pool. Under the method above, the all-in fee puts the network line on the vendor’s side.

3. Self-hosted open-source code. No service fee, and for a technically capable operator running repeated campaigns, likely the cheapest route. You carry every network line, the engineering, and the maintenance when a DEX program changes. What public repositories give you and what they leave out is covered in open-source volume bots.

How the first two price the same 100 SOL session, line by line, is worked through in flat-fee versus per-trade pricing.

Which features actually matter

Four do real work: automatic handoff at graduation, so activity does not stall during migration; accurate compute limit handling, which controls fee waste; landed-swap accounting rather than attempt counting; and configurable duration, which controls self-contention. Most other feature-list entries are either table stakes or decoration.

Graduation handoff. Migration is when new attention arrives. A campaign that stops at the handoff goes flat at the worst moment. Detail in Raydium routing.
Compute limit accuracy. Padding limits raises cost at the same bid; undershooting causes failures that still charge. This is invisible in a feature list and directly visible in the bill. See priority fees.
Landed-swap accounting. If a dashboard reports attempts, you cannot tell whether a campaign worked. This is the difference between reporting and telemetry.
Duration control. A large fleet compressed into a short window contends with itself and raises its own failure rate. Covered in wallet count.

Features that sound impressive and mostly are not: proprietary algorithm names, undisclosed "anti-detection" claims, and dashboards showing volume that cannot be reconciled against chain data. The first is branding, the second is examined in clustering risk, and the third is a reason to leave.

Where Volion sits, stated plainly

Volion charges a flat 2% of target volume, is non-custodial, routes to the Pump.fun curve, the Pump.fun AMM or Raydium with automatic handoff, and publishes its measured cost and failure data with the sampling method attached. It is not the cheapest option available, and for a technically capable operator running repeated campaigns, self-hosting public code is likely cheaper.

That last sentence is not modesty, it is the honest read of the arithmetic in free volume bots. The service fee buys the operational layer: fee tuning, retry logic, RPC capacity, accounting, and maintenance when a DEX program changes. If you intend to own that layer yourself and can, you should keep the fee.

What we would argue distinguishes this tool is the measurement discipline rather than any feature. The failure rates are published because they change the budget, the method is published so the figures can be checked, and the limits are stated on their own pages rather than buried. Whether that is worth a percentage is a judgement you are better placed to make than we are.

What we will not claim: that volume produces demand, that trending can be guaranteed, or that a fleet is undetectable. Those three claims are common in this market and none of them survive contact with the chain.

If you are ready to compare totals rather than adjectives, the measured figures behind everything above are on the Solana volume bot overview, the fee is on the pricing page, and the arithmetic that tests any competing quote is on the cost floor page.

Add one more question to any comparison: which venues does the tool actually route to, and does it follow a token across them? A bot that handles Raydium but not the Pump.fun curve is useless before graduation; one that handles the curve but not PumpSwap goes silent at exactly the wrong moment. Full coverage today means the curve, PumpSwap, Raydium AMM and CLMM, Meteora DLMM, Orca Whirlpools and a considered position on Jupiter routing.

Questions

What is the best Solana volume bot?

There is no answer that is both honest and generic, because the right choice depends on whether you can self-host, how many campaigns you intend to run, and which routing target you need. What is transferable is the comparison method: convert every quote into cost per landed swap on your target, and ask each vendor what their tool cannot do.

Why is your short list so short?

Because any ranking published by a vendor is written by someone with a stake in the outcome, and you should discount ours accordingly. We list only options whose pricing and custody model are published clearly enough to run the comparison method against: Volion, PumpWave, and self-hosted open-source code.

What is cost per landed swap?

The fee actually paid divided by the success rate of the routing target. It is the only figure that makes two quotes comparable, because a vendor quoting a price per attempt is quoting something that does not equal delivered volume.

What features actually matter?

Automatic handoff at graduation, accurate compute limit handling, landed-swap accounting rather than attempt counting, and configurable duration. Most other feature-list entries are either standard across all tools or decorative.

Is Volion the cheapest?

No. For a technically capable operator running repeated campaigns, self-hosting public code is likely cheaper, and we say so on the free volume bots page. The service fee covers the operational layer rather than the software itself.

Measured transaction costs Cost breakdown Free volume bots Open-source volume bots Is a volume bot safe Flat fee vs per-trade fees