AUTHORIZED L4 + L7 · EXECUTION-ONLY
Pressure, on demand.
fluxxy is an authorized network stress and load testing operator. Twenty-four methods across layer 7 and layer 4, capacity provisioned before you sign in, sustained runs to 3,600 seconds — and a report for every run.
// PREPAID · ONE-TIME · STAFF-CONFIRMED · NO SUBSCRIPTIONS
// LIVE FIGURES ONLY · “—” MEANS THE BOARD IS UNREACHABLE · NOTHING ON THIS PAGE IS A PLACEHOLDER NUMBER
The console
REF STANDBY DEMOOne surface: pick a method, size the window, queue it — and the chunk chain holds the line without you watching. This frame is a standby demo: the SIM beat is staged, the slot numerals are not.
One invocation
REF EV-RUNA launch from the console side: request in, run in flight, report out. Both panels below are staged — the band above is not.
12:04:53 boot l7-http 512c 12:04:54 ramp 0 → 512 conc over 8 s 12:05:31 steady 41,200 req/s p99 182 ms 12:06:12 chain chunk 2/5 queued 12:09:53 complete 300 s · 0 aborts
Method arsenal
REF 24 METHODSTwenty-four methods, four families, one chaining mode. Everything below is a live selector in the console — no rounding up, no “and more”.
Smart layer-7 floods (http) plus TLS handshake pressure (tls). Requests behave like traffic, not like a script — headers, pacing and connection behaviour included.
METHODS http · tls / SOURCE REAL / RUN 3,600s
Raw TCP and UDP legs: connection floods, spoofed-source SYN and ACK where the route allows it, pipe-filling data pressure and datagram volume.
METHODS tcp · syn · ack · tcpdata · udp / SOURCE REAL · SPOOFED
Legs that speak the protocol instead of brute-forcing the port — Source engine, Rust, TeamSpeak, RakNet and the GTA multiplayer family.
METHODS valve · rust · ts3 · raknet · samp · mta · fivem / UDP FACING
For the edges the tidy families don't cover: GRE tunnels, ICMP pressure, OpenVPN endpoints and randomized source rotation.
METHODS gre · icmp · ovpn · iprand / LAYER 3 – 4
Any run past 300 s chains itself in minute chunks to a 3,600 s ceiling. Queue it once; the pad holds the line until the run completes.
CHAIN AUTO / CHUNK 60s / CEILING 3,600s
Allocations
REF FX-1 / FX-4 / FX-16Three standard builds. Prepaid and one-time — you agree the build with staff, pay, and the grant activates when staff confirm it.
| REF | CONCURRENTS | RUNTIME CEILING | API | PRICE | Reserve |
|---|---|---|---|---|---|
| FX-1 | 1CONC | 300s | +$10/mo | $15 | Reserve |
| FX-4 | 4CONC | 900s | +$10/mo | $43.50 | Reserve |
| FX-16 | 16CONC | 1,800s | +$10/mo | $157.50 | Reserve |
Illustrative rhythm — queue a window and staff confirm the slot.
- Prepaid, one-time. No subscriptions, no stored cards.
- Every build runs all twenty-four methods — L7, L4 and game — nothing is gated behind a tier.
- A human confirms every grant before it goes live — no instant card-to-firepool.
- Slots are held per launch window; queue a window and your concurrents sit locked until T-0.
- Custom builds and runtimes past the standard ceilings are quoted on Telegram.
// PREPAID · ONE-TIME · ACTIVATES WHEN STAFF CONFIRM — CUSTOM BUILDS ARE QUOTED AT t.me/user27124
Why fluxxy
REF POSTUREA categorical comparison, not a benchmark — how we operate against how subscription stress vendors typically do.
// CATEGORICAL, NOT MEASURED — NO PROVIDER FIGURES APPEAR HERE BECAUSE NONE WERE MEASURED
API
REF API-V1The console is an API with a UI on it. Bearer keys, one endpoint per action, JSON out — $10/month as an add-on to any build. Same methods, same limits, no separate infrastructure.
curl -X POST https://<panel>/api/benchmark \ -H "Authorization: Bearer fxk_YOUR_KEY" \ -d method=http \ -d target=example.invalid \ -d concurrents=512 \ -d time=300
{
"ok": true,
"run": "RUN-04821",
"queued": false,
"method": "http",
"concurrents": 512,
"time": 300,
"state": "in_flight"
}
Full request shape and error codes at /apidocs.
Ask the console
REF VEXAVexa sits in the console and answers in plain language — methods, allocations, why a run throttled. The real chat opens after you sign in; the exchange below is staged.
Which method fits a big HTTP target in 10 minutes?
vexa> l7-http at 512 concurrents, 600 s. Past 300 s it chains minute chunks on its own — nothing to babysit.
Why did my last run throttle?
vexa> Your concurrent slots were busy in that window. Queue the run for a quieter one, or raise concurrents on the build.
What does an allocation include?
vexa> Build your own: $15 base covers 1 concurrent + 300 s, +$9.50 per extra concurrent, +$0.50 per extra minute past 300 s. API is +$10/month. Prepaid, one-time, via staff.
How we operate
REF POLICY- Uptime target 99.98%, measured on the pad, not the brochure.
- Authorized targets only. Every engagement assumes you own the target or hold written permission.
- Staff-confirmed provisioning. A human confirms every grant before it goes live — no instant card-to-firepool.
- Ticket support on Telegram. Staff answer fast, and there is no queue theatre.
// EXECUTION-ONLY SERVICE · AUTHORIZED TARGETS ONLY · VIOLATIONS END THE ENGAGEMENT AND THE ACCOUNT
Sign in, open /purchase and move the slider. The $15 base covers one concurrent and 300 s; each extra concurrent is +$9.50 and each extra minute past 300 s is +$0.50. Longer runs and custom builds are quoted on Telegram.
Builds are prepaid and one-time. You agree the build with staff on Telegram, pay, and the grant activates when staff confirm it. No subscriptions and no stored cards.
Anything past 300 s chains automatically in minute chunks, up to a 3,600 s ceiling. You queue it once and the pad holds the line until the run completes.
Yes. Queue a window and your concurrents are held until T-0. The launch fires whether or not you are watching the console, and the report is waiting after.
Yes — REST with bearer keys, $10/month as an add-on. The API fires the same methods as the console; /apidocs has the full request shape.
Targets you own or are authorized to test, and nothing else. Unauthorized targets end the engagement and the account. One-time prepaid builds — if you're unsure, talk to staff on Telegram before reserving.