Solana sniper bots

The fee stack that decides everything

Bot	Type	Keys	Price	Trade fee	Free tier	US	Status
AX Axiom	DEX / Telegram sniper	wallet	0.7% per trade	0.7% per trade	yes	-	-
Banana Gun	DEX / Telegram sniper	wallet	1% per trade	1% per trade	yes	-	-
BullX	DEX / Telegram sniper	wallet	1% per trade	1% per trade	yes	-	-
PH Photon	DEX / Telegram sniper	wallet	0.9% per trade	0.9% per trade	yes	-	-
GM GMGN	DEX / Telegram sniper	wallet	1% per trade	1% per trade	yes	-	-
Maestro	DEX / Telegram sniper	wallet	1% per trade	1% per trade	yes	-	-
TR Trojan	DEX / Telegram sniper	wallet	0.9% per trade	0.9% per trade	yes	-	-
BO BONKbot	DEX / Telegram sniper	wallet	1% per trade	1% per trade	yes	-	-

Live values from our database - details and caveats in each review. "-" = not documented.

The category's economics are unforgiving by construction. A "free to install" Telegram bot charging 0.9-1% per side needs a token to move ~5% in your favor per round trip just to break even - before taxes, before the launches that never let you sell, before the rug. Compare: an exchange-native grid bot pays 0.05% per side, twenty times less, with a real counterparty (our grid hub does that fee math in full).

Across the 8 snipers we track, the base fee runs 0.7%-1% - and 8 of them are wallet-connected, meaning your approvals are the custody surface[8]. The transparency signal is worth reading before the fee: Banana Gun documents its charge on a fetchable site[1]; BullX and Maestro serve JS shells[2][3]; Photon and Axiom block datacenter verification entirely[4][5]. In a lane where the operator holds your trading flow, opacity is data.

The rug reality, stated plainly

Memecoin launches on permissionless DEXs are adversarial by design: mint authority tricks, LP manipulations, bundled supply, coordinated dumps. The crime literature treats the rug as the structural default of unvetted launches[9]; our scam patterns guide adds the follow-on layer - fake recovery services that re-victimize. Anti-rug filters (mint revoked, LP burned, holder distribution) filter yesterday's rug; the competition you are sniping against upgrades first and faster.

This is why our scores in this lane run 5.0-5.7 - the lowest on the site - not because the bots malfunction, but because the game's expected value is negative for most participants and the platforms monetize participation, not outcomes. The FCA's bot-branded warnings target the same audience through the same channels (Instagram, Telegram)[8]; the scam funnel documented in our scam patterns guide starts here.

The referral economy, decoded

Almost every sniper in this table runs a two-way referral system: invite codes grant fee discounts, referrers earn a share of the referred users' fees. That explains the shape of the community corpus - the highest-scoring posts about these bots are referral-linked filter guides[6][7], and every fee number arrives with a discount code attached. Referrals are not inherently dishonest; they are structurally biased - the referrer profits from your trading volume, not your outcomes. Read every fee claim twice: once against the operator's own documentation where it exists[1], once against your own transaction history after the first trades.

The referral economics also explain feature churn. Filter presets, speed tiers and "premium" RPC access are differentiation levers aimed at referrer promotion cycles as much as at utility. When a bot's most-discussed feature is its invite program rather than its fill quality, that is the market telling you what actually sells.

If you participate anyway - rules that contain the damage

• Predefine the loss budget. An entertainment budget, in writing, per month. The lane's failure mode is the top-up; the position size calculator formalizes what you will not exceed.
• Revoke approvals after every session. Wallet-connect bots need approvals to trade; standing approvals to unknown routers are open doors[9]. A revocation routine is the cheapest security in crypto.
• Treat referral-linked fee claims as marketing. The community corpus is dominated by referral posts with discount codes[6][7] - cross-check every fee number against the operator's own documentation where it exists[1].
• Sell rules matter more than buy filters. Auto-sell on liquidity removal is a feature worth paying for; manual exits during a rug are exits at the bottom.
• Never reuse a trading wallet with real holdings. One wallet for the lane, funded with its budget, nothing else in it. A drained sniping wallet is a bounded loss; a drained main wallet is a life event.

Warnings

• This is the only lane on our site with an average score below 6 - documented on purpose. The bots work; the game is structured against participants[9].
• "Promised pump" channels and paid alpha groups are the scam layer on top of the scam layer - the FTC's patterns apply verbatim, with the FCA listing bot-branded unauthorised firms[8]. Report, block, move on.
• Operator opacity is a risk signal, not an inconvenience. JS-shell sites and 403-blocked fee pages[2][4][5] mean the fee table you act on is community-documented - verify against your own transaction history and dispute nothing you cannot prove.
• Priority fees and tips spike at launches - the exact moments you trade. A fee model based on quiet-market numbers understates the hurdle you will actually pay.
• Jurisdiction: none of these operators document US availability in our tracking. If that matters to you, the US-available lane is the compliance-respecting alternative.

The lane in numbers

• 8 snipers are tracked in this lane; 8 of them connect by wallet rather than exchange keys - approvals are the custody surface, not key vaults[8].
• 0 offer a formal trial - effectively none of the lane does, which makes every first deposit a leap; size accordingly.
• 5.3 is the median overall score here - the lowest of any lane on this site, reflecting documented transparency gaps and the structural loss profile[9].
• Across the whole tracked universe, 2 platforms have gone under and 0 sit on our watch list - the churn rate in fast-moving lanes is itself a risk signal.

Comparison honesty in a lane that resists comparison

Most of this page's bots resist the verification our other lanes get: operator sites that block datacenter IPs[4][5], JS-only rendering[2][3], fee pages that live in referral posts[6][7]. We document each fetch result honestly rather than guessing - and our reviews carry the per-bot transparency notes. That is why the fee table above cites community documentation where operator pages fail: it is the honest state of what can be verified, marked as such, not a clean matrix pretending otherwise.

What can be verified universally is the math: base fees are charged on-chain per trade and visible in your own transaction history. After your first ten trades, reconcile every fee line against this page's table - discrepancies between marketed and charged fees are exactly the kind of finding our comment sections exist for. The lane's information vacuum is filled by users who check their own fills - not by users who hope. Reconcile early, reconcile often, and the marketed fee table stops being a stranger.

A closing calibration: every other lane on this site gets a recommendation-shaped summary because its trade-offs are priceable. This one gets a warning-shaped one instead - the bots compete hard, the fees are knowable, the filters improve every quarter, and the game still takes most of its participants' money to the structure of launches themselves[9]. If that paragraph reads as a dare, budget it like one - in writing, with a monthly ceiling, before the first Telegram bot ever opens.