This build of VANTYX:Server-authoritative21 in-game detectionsBlocks in-tickWorks offline0 automated bans0 client-side sensors
NOTHING GETSTHROUGH.
Server-side FiveM anti-cheat that blocks the damage in-tick on your own server, keeps protecting when the cloud is down, explains every flag in plain English, and never bans anyone by itself.
Detection engines
The damage is stopped where it happens, not reported after it.
VANTYX installs as one FiveM resource on your server. There is no client-side download, no launcher and no driver: your players install nothing, and nothing of ours runs on their machine. Every check reads state your server already holds, or an event your server is already routing — which is why blocking happens in the same tick, and why it keeps happening when our backend is unreachable.
When the connection comes back, the findings upload and the dashboard explains each one in plain English: what happened, why it looked wrong, what evidence sits behind it, and which innocent explanations were considered and ruled out. Nobody is banned by that process. It ends with a proposal and a human.
- 21
- checks that run in Lua on your own FXServer. They cancel the event in the same tick it arrives, before the money moves or the explosion happens.
- 3
- checks that run in our backend, where a pattern needs a window rather than a tick. They produce findings; they never block.
- 59
- written explanations of how a check fires on somebody who did nothing wrong — two per detection, minimum, written before any of them ever fired.
- 0
- automated bans. The most severe thing VANTYX does on its own: “proposed a temporary ban for a staff member to review.”
Every check, in the words your dashboard will use for it.
This is the catalog itself rather than a summary of it. Pick an engine to see what ships behind it: each name and description below is the exact string a server owner reads when that check fires, and the badge under it is the most severe thing the check is ever allowed to cause on its own.
Entities
Runs on your server · blocks in-tick · keeps working offline
Invisible Explosion
A player created an explosion with no visual effect but real damage.
Blocked Explosion Type
A player created an explosion of a type your server does not allow.
Amplified Explosion
A player created an explosion with damage turned up beyond normal.
Remote Explosion
An explosion attributed to a player happened far away from them and from anything they were controlling.
Blocked Vehicle or Object
A player tried to spawn something your server does not allow.
Rapid Spawning
A player created an unusual number of entities in a few seconds.
State Flooding
A player wrote to networked state far faster than normal.
Three lines in server.cfg, and nothing for your players to download.
About ten minutes, and you do not edit any Lua. Before you ever open the dashboard, the console tells you what VANTYX can and cannot do on your particular server — every check it had to switch off is named, with the reason and the fix.
Drop in the resource
Copy the
vantyxfolder intoresources. There is no second download for your players, and nothing to distribute to them.Add three lines to server.cfg
ensure vantyxat the top of your resource list — it can only protect events that were registered after it started — then your key and a folder it can write to.Restart and read the console
It reports what it can and cannot do on your server, in words, with the fix attached to each problem.
vantyx checkupprints that report again at any time.
ensure vantyx
set vantyx_license_key "vtx_live_…"
set vantyx_state_path "/home/fivem/server-data/.vantyx/"
NOTHING GETS THROUGH.
Type vantyx status for a health summary.
[VANTYX] Environment check passed. All 21 detections available.
[VANTYX] Everything is up. Type vantyx status at any time.There is no VANTYX on your players’ machines. Including where that costs us.
Most anti-cheats ask a player to run something. VANTYX asks for nothing, because the client sensor tier was cut from the design rather than postponed — every shipped check reads what your server already knows. That removes a whole class of bypass, and it leaves one thing genuinely out of reach.
| Question | VANTYX — server-side only | A client-side sensor |
|---|---|---|
| What runs on the player’s machine | Nothing at all. There is no client component, so there is nothing to install, nothing to crash, and nothing to blame for a frame-time drop. | A module inside the game process, which the player has to run and trust, and which the cheat runs alongside. |
| What a new cheat build has to defeat | Your server’s own view of what happened. The check is not on the machine the cheat controls, so there is nothing on that side to patch out. | The sensor. Once a build evades it, coverage is gone until the vendor ships a new one. |
| What happens when the vendor’s backend is unreachable | Every in-game check keeps running and keeps blocking. Findings spool to disk and upload when the connection returns. | Varies by vendor. |
| Money duplication through server events | This is the strongest surface: VANTYX owns the event registration, validates the payload shape and can refuse the event outright. | Not what a memory scanner is looking at. A duplicated item is a legitimate-looking event from a compromised client. |
| An external cheat that only READS memory (ESP, radar) | Honest limit: Invisible. It touches no event, desyncs no position and asks the server for nothing. No server-side approach can see it, and VANTYX does not claim to. | Sometimes caught, if the sensor is present, current, and not itself bypassed. |
| What it collects about your players | What the server already knows: their in-game actions. No hardware identifiers, no screen contents, no process list. | Whatever the module is built to read on the machine it runs on. Hardware identifiers are the usual pitch. |
The things it will not do, and what each refusal costs you.
Every item below is something a competing product advertises. They are positions rather than gaps, so each one is stated with the price it carries — a refusal with no cost attached is marketing, and you would find the cost on your own server anyway.
No hardware fingerprinting
Nothing runs on the player's machine: no hardware fingerprinting, no screen capture, no memory or process scanning. VANTYX protects servers; it does not profile the machines of the people connecting to them.
The cost: ban evasion is cheaper here than on a product that fingerprints hardware. A banned player who buys a new account can come back, and we surface linked accounts as a signal for a person to weigh rather than as grounds on their own.
No screen capture, ever
No screenshots, no overlay capture, no window titles. Nothing about what is on a player’s screen reaches us, because nothing of ours is on their machine to look.
The cost: a cheat’s on-screen overlay is not evidence we can show you. What we can show you is what it made the server do.
No memory or process scanning
We do not read game memory, enumerate processes or list files. That is what an anti-cheat rootkit does, and it is the reason players distrust them.
The cost, stated in the comparison above: an external cheat that only reads memory is invisible to us.
No shared ban network
Your players’ records are yours. Tenant isolation is a hard boundary in the database itself, and identifiers are hashed with a key derived per organisation, so one customer’s data cannot be joined against another’s even with a full dump.
The cost: a cheater banned on somebody else’s server arrives on yours unknown, and you judge them on what they do to you.
No automated bans
The enforcement ladder stops before a ban. The most severe thing VANTYX does by itself is: “proposed a temporary ban for a staff member to review.” A person bans, from the dashboard, with the evidence bundle attached — the database will not store a ban without one.
The cost: somebody has to look. In exchange, no configuration mistake of ours can empty your server overnight.
No AI deciding anything
There is no model in the enforcement path, and none runs at all unless your organisation turns one on. The explanation you read on a flag is assembled from the catalog’s own wording and the actual facts of the event — a template, not a paraphrase.
The cost: the explanations read the same way every time. We consider that a feature in a document somebody may have to defend.
No learning loop pretending to be trained
The closed adaptation loop is built and switched off, at two independent levels, until there is labelled data from real servers to calibrate it against. What is live instead: shipped starting profiles per server type, hard physics ceilings nothing may cross, an SDK your own scripts use to authorise their own actions, and threshold changes a human approves.
The cost: your tuning improves when you approve a change, not silently overnight. A confidently wrong threshold is worse than a conservative one.
One resource. Per server. Every detection included.
No tier that unlocks the checks that matter, and no per-player metering to make a busy night expensive.
These amounts are placeholders. VANTYX has not set its prices. We would rather show you a marked placeholder than a number that moves before launch — everything else on this page is settled.
Starter
One server, one community.
€29 / month placeholder
- Servers
- 1 server included
- Extra servers
- Not sold separately on this plan.
- History
- 365 days of detections. Every plan keeps the same history today. Retention is uniform across plans and the per-plan tiering is not implemented, so we do not sell it.
- Support
- Email support.
- AI Investigator
- Not on this plan. Every case still comes with the full deterministic explanation — what happened, why it looked wrong, and the legitimate causes we considered — because that is the same text the Investigator is grounded in.
Pro
A few servers under one roof, with seats you can add as you open them.
€79 / month placeholder
- Servers
- 3 servers included
- Extra servers
- €24 each per month — placeholder.
- History
- 365 days of detections. Every plan keeps the same history today. Retention is uniform across plans and the per-plan tiering is not implemented, so we do not sell it.
- Support
- Email support.
- AI Investigator
- Available. Off until you turn it on. It reads a case that has already been assembled and writes the summary in plain English; it cannot ban, kick, or change a setting, and switching it off returns you to the deterministic explanation you get anyway.
Enterprise
A network large enough that the seat count and the support arrangement are a conversation.
Quoted placeholder
- Servers
- 25 servers included
- Extra servers
- Not sold separately on this plan.
- History
- 365 days of detections. Every plan keeps the same history today. Retention is uniform across plans and the per-plan tiering is not implemented, so we do not sell it.
- Support
- Priority support, with an agreed response time.
- AI Investigator
- Available. Off until you turn it on. It reads a case that has already been assembled and writes the summary in plain English; it cannot ban, kick, or change a setting, and switching it off returns you to the deterministic explanation you get anyway.
In every plan, including the cheapest
- Every detection in the catalog. No tier withholds a check.
- The dashboard: player records, evidence, plain-English explanations, configuration.
- The integration SDK, so your own scripts authorise their own actions instead of being flagged for them.
- Protection that keeps running with our service unreachable.
- A lapsed licence that never stops the resource and never drops a player.