Nodetonet is prepaid. There is no monthly subscription, no overage invoice, and no long-term contract. Instead you top up service credit and the platform debits it once per day for every proxy that was actively running at the settlement moment. You watch the balance tick down at a predictable rate. This post is the plain-English explanation of exactly how that meter works.
In one sentence: running proxy = a small daily charge; stopped proxy = zero cost. Two rules govern everything.
The billing unit: one active tunnel, one day
The atomic billing unit is a proxy (tunnel) that was running at the moment of the daily settlement. A proxy is "running" if it has a live listener on an edge server and can accept connections — whether it actually carried any traffic that day does not affect the price. We bill capacity, not bytes.
The price for one tunnel-day is currently ~$0.033. That works out to roughly $1 per tunnel per month, which is why the smallest top-up package is exactly $1 — it gives you a comfortable month to validate the platform with a single proxy before committing more.
How cost scales with the number of running proxies:
| Running proxies | Daily cost | 30-day estimate |
|---|---|---|
| 1 | ~$0.033 | ~$1 |
| 5 | ~$0.165 | ~$5 |
| 10 | ~$0.33 | ~$10 |
| 50 | ~$1.65 | ~$50 |
Idle (stopped) proxies appear in none of these rows. They carry zero cost.
03:00 UTC: the settlement window
Once every UTC day, at exactly 03:00, a scheduled job sweeps across every account. It asks each proxy "are you running right now?" and debits one day's price for every yes. The entire sweep finishes in seconds; every account ledger is current by 03:00:30.
The practical implications are worth spelling out carefully:
- A tunnel that ran for 23 hours and was stopped at 02:59 UTC is not billed for that day. Settlement only sees the state at the instant 03:00 fires.
- Conversely, a tunnel started at 02:55 and stopped at 03:05 is billed for that day. Don't try to game it — just leave it running when you need it.
- Time zones matter only for your mental model. The job is always UTC. If you're in UTC+3 the daily charge lands at roughly 06:00 local time.
Idle tunnels burn nothing
The most important rule for cost control: stopped tunnels cost nothing at all. Not a fractional cent, not a "reservation" fee — nothing. You can sit on a hundred configured-but-stopped tunnels for a year and pay $0 for the privilege. The configuration itself — token binding, protocol choice, port assignment, ACLs — is free. Only the active listener has a cost.
For unpredictable workloads, the natural pattern is to leave proxies stopped between bursts and bulk-start them when you need them. Bulk operations covers the mechanics; the key point here is that the pricing model actively rewards this pattern.
Where to read the ledger
The full ledger lives at /balance, in the Transaction history table at the bottom of the page. Every row contains:
- Timestamp — when the row was written, in UTC. Daily debits all share the 03:00 timestamp; top-ups land whenever Stripe's webhook fires.
- Type —
credit(top-up, bonus, refund),debit(daily settlement), oradjustment(manual support correction). - Amount — signed. Debits are negative; top-ups are positive.
- Reference — for top-ups, a link to the Stripe receipt. For debits, the count of tunnels charged that day.
The dashboard "service credit" tile shows the headline number, but it also shows a "~N days at current burn" estimate. That estimate uses your last 7 daily debits as the rolling burn rate. If you have auto top-up disabled, use this number to plan when your next top-up needs to land.
What happens when credit runs out
If a daily settlement would debit more than your remaining balance, the platform does not put you into debt. Instead, every running tunnel is gracefully stopped within seconds of the failed debit, and the account flips to a "credit exhausted" banner. Your configurations remain intact — top up any amount and bulk-start your proxies back up. No data is lost.
If you never want to encounter this state, save a card and enable auto top-up. The platform charges a fixed amount whenever the balance dips below your chosen floor, with a ten-minute cooldown to prevent runaway double-charges.
Nodetonet pricing vs per-GB billing — a quick comparison
Many competing providers sell residential proxy capacity by the gigabyte, which means costs scale directly with traffic volume and can spike unexpectedly during heavy crawls. Nodetonet's per-device/per-day model puts a ceiling on what any given day can cost, regardless of how much traffic you push through. Whether a proxy moves 10 GB or 200 GB in a day, you pay the same daily unit. For a worked cost comparison see pay-as-you-go pricing.
If you're not yet sure which proxy type fits your use case, start with what is a mobile proxy, then explore the mobile proxies feature page.
Practical tips for keeping costs low
- Use stop/start cycles for burst workloads. When a task finishes, stop the proxies via the panel or the API. You pay nothing until the next run.
- Pool devices into token groups. With rotating proxies, a single token group can round-robin across multiple devices, so you don't need all of them running simultaneously.
- Set per-client quotas. The proxy clients feature lets you cap bandwidth per customer or use-case, keeping individual usage predictable.
- Check the burn-rate estimate weekly. The "~N days" figure on the dashboard surfaces unexpected load before it drains your balance.
- Keep auto top-up active for always-on proxies. If any of your proxies run 24/7, a configured auto top-up is the simplest way to avoid an unplanned service interruption.
What's next
- Saving cards and auto top-up — never get stopped out unexpectedly.
- The Stripe top-up flow — what happens between Pay and Balance Updated.
- Refund policy — what's refundable, what isn't.
- Bulk operations — start or stop dozens of proxies at once.
- The quickstart guide — for the wider context of what those proxies do.