Best Pump.fun volume bot settings for trending
The best Pump.fun volume bot settings spread activity across many wallets (hundreds for a real trending push, not a handful), randomize per-trade SOL size instead of using fixed amounts, run a buy-weighted-but-not-pure-buy mix, and time the session to start with your announcement - while auto-comments, auto-favorites and anti-MEV Jito routing stay tuned to volume automatically. This is a practical, knob-by-knob playbook for configuring a session that reads as a genuine crowd rather than an obvious script.
Wallet count: the diversity dial
Wallet count is the first and most consequential setting, because it decides whether your activity looks like a crowd or like one address talking to itself. Every distinct wallet that buys and holds adds to the token's holder count, and holder growth is one of the clearest signals that a launch is catching on. If you run volume through only a few wallets, the tape can look busy while the holder number barely moves - a mismatch that experienced traders spot instantly.
The working principle: scale wallet count with the size of the push. A quick test to see the mechanics can run on a few dozen wallets. A launch you are seriously trying to trend wants activity spread across hundreds of independent wallets, so that buys genuinely lift the holder count and no single address ever dominates the flow. The wallet count guide goes deeper on choosing a figure for your budget.
Per-trade SOL size
The rule for trade size is: never use a fixed number. A tape where every buy is exactly 0.1 SOL is a signature no amount of other tuning can hide. A good configuration randomizes each trade's size within a range, so the flow has the natural variance of real traders making different-sized bets. On the early bonding curve, where the pool is shallow, even small trades move price and register as activity, so you do not need large sizes to make an impression. After the token graduates to a Raydium pool, the calculus flips: trade size has to stay modest relative to the pool's depth, or you eat slippage and paint an obvious target for sandwich bots. The engine shifts sizing behavior to fit the phase, so you set the range and it handles the adaptation.
Session duration
Duration should be chosen against a goal, not picked at random. Two patterns cover most launches. The first is a short, dense burst: high activity concentrated into the minutes around your announcement, designed to shove a fresh token onto the trending feed while the most eyes are scanning new launches. The second is a longer, steadier session that carries a token across its whole bonding curve and into graduation, so it never goes quiet during the climb. Many launches combine them - open with an intense burst to break through, then drop to a lower-intensity tail that holds visibility without burning capital. Because the dashboard lets you set duration directly, you are not forced into a fixed package.
Buy pressure
Buy pressure is the ratio of buys to sells the engine aims for, and it is how you control whether the token looks like it is accumulating or just churning. During a trending push you generally want a buy-weighted mix: more buys than sells pushes the token up the curve and signals that participants are accumulating rather than dumping. But pure buying is a mistake - a pool that literally never sells looks artificial and traders read it as a trap. The healthy default is buy-weighted with a genuine minority of sells mixed in, which mirrors how a real early crowd behaves. Tighten the skew when you want a harder climb; loosen it when you want to look like organic two-way trading rather than a directed push.
Auto-comments and auto-favorites
Here is the setting most people over-think: auto-comment and auto-favorite density is not something you dial in as a raw count. It is auto-tuned to volume. The engine keeps the social surface proportionate to the trading activity, because the failure mode is a mismatch - a token showing heavy volume but a dead comment thread looks stranger than one with no comments at all. By scaling discussion and reactions to the tape, the page stays internally consistent: busy trading, busy conversation. Two things to keep in mind. First, these features apply to Pump.fun and Bonk.fun tokens, both of which are social launchpads where the page itself is part of the pitch; a plain Raydium pair has no comment thread to fill. Second, because it is automatic, you do not risk the classic error of blasting fifty comments onto a page with three trades.
Timing, curve and the anti-MEV layer
Two settings here run automatically but are worth understanding. The first is timing: the engine spaces trades on a human-like, Poisson-style distribution rather than a fixed interval, because evenly spaced trades every N seconds are a dead giveaway. Real trading is bursty and irregular, and the engine imitates that. The second is anti-MEV routing, which is on by design - every order goes through Jito bundles with randomized tips so it is not exposed on the public mempool where sandwich bots wait. This matters most after graduation, when the token trades on a Raydium AMM and MEV bots actively hunt large or predictable orders. Optional cross-DEX mirroring across Meteora and Orca can widen your footprint on aggregators like Dexscreener and Dextools. For the mechanics of the curve itself, see the bonding curve explainer, and for the curve-to-Raydium handoff, the Raydium volume bot page.
When to launch relative to your announcement
The most common timing mistake is starting the bot after the announcement instead of with it. The whole point is that your earliest visitors - the ones who click within seconds of you posting - land on a page that is already alive: live volume, holder count ticking up, comments moving. If they instead find a silent page because you are still configuring a session, you have spent your best traffic on your worst impression, and first impressions on Pump.fun do not come back. Configure the session in advance and start it just before or exactly as your announcement goes out. For the wider first-hour strategy this settings work supports, see the trending playbook. And remember the honest limit: good settings give a launch its fair shot at attention, but they cannot manufacture demand for a token nobody wants.
Frequently asked questions
What is the best wallet count for a Pump.fun volume bot?
There is no universal number, but the principle is diversity: more wallets means holder growth looks organic and no single address dominates the tape. A small test might use a few dozen wallets, while a serious trending push spreads activity across hundreds so that buys translate into a rising holder count rather than one wallet churning the same balance. The right figure scales with your budget and how long the session runs.
How much SOL should each trade move?
Per-trade size should be varied, not fixed. A tape where every buy is the same round number is an obvious script, so a good configuration randomizes trade size within a range. On the early bonding curve small trades are enough to move the needle because the pool is shallow; after graduation to Raydium, size has to stay modest relative to pool depth to avoid slippage and sandwich bots. The engine adapts sizing to the phase automatically.
How long should a session last?
Match duration to your goal. A short, dense burst timed with an announcement helps a fresh launch clear its first quiet minutes and hit the trending feed. A longer, steadier session sustains a token through the full curve and into its Raydium graduation. Many launches use a strong opening burst followed by a lower-intensity tail to hold visibility without burning through capital.
What is buy pressure and how should I set it?
Buy pressure is the ratio of buys to sells the engine targets. A higher buy skew pushes the token up the curve and signals accumulation, which is what you usually want during a trending push. Some sells are still healthy because a pool that only ever buys looks unnatural. A balanced-but-buy-weighted mix is the typical starting point, tightened or loosened depending on how aggressively you want to climb.
Are auto-comments and auto-favorites something I tune manually?
They are auto-tuned to volume rather than set as raw counts. The engine keeps comment and favorite density proportionate to the trading activity so the social surface always matches the tape - a token showing heavy volume but an empty comment thread would look wrong. These features apply to Pump.fun and Bonk.fun tokens, which are social launchpads; a plain Raydium pair has no comment thread to populate.
Do I need to configure anti-MEV routing?
No manual setup is required - anti-MEV Jito routing is on by design. Every order is submitted through Jito bundles with randomized tips so that large or predictable trades are not picked off by sandwich bots on the public mempool. This matters most after graduation, when the token trades on a Raydium AMM where MEV bots actively hunt exploitable orders.
When should I launch the bot relative to my announcement?
Start the session slightly before or exactly as your announcement goes out, not long after. The goal is for the first wave of real visitors to arrive at a page that already shows live volume, growing holders and active discussion. A token that is announced and then sits silent for ten minutes while you set up the bot has already lost the impression, because the earliest viewers saw an empty page.
Can one settings profile carry from the curve to Raydium?
Yes. You configure once, and the engine detects graduation and re-points execution into the Raydium pool without a gap, adjusting sizing and routing for the AMM automatically. You do not need a second tool or a second configuration for the post-graduation phase, and the flat 2% commission covers both halves of the run.
Good settings come down to one idea: make the activity look like a real crowd, and make every surface of the page agree with every other. Diverse wallets, varied trade sizes, buy-weighted flow, human timing and social density scaled to volume all serve that single goal. Configure a session and put it to work.