How LocalProxies Pricing Works

If you are comparing proxy providers, the pricing pages rarely line up: some quote per gigabyte, some per IP, some per request. This guide explains exactly how LocalProxies prices residential proxies, so you can size a plan and predict your bill with no surprises.

The short version: you pay for bandwidth. LocalProxies charges for the one input that actually scales with cost on a residential network, the data that flows, and provisions it through a concurrency tier. The billing unit is the gigabyte, not the IP or the request.

What you pay for: bandwidth

LocalProxies bills for total data transfer (upload and download combined), metered per gigabyte, at a rate that falls as your volume grows, from $3.00/GB on entry tiers to $2.10/GB on the largest tiers.

Two plan families: Ports and Threads

You do not buy raw bandwidth in the abstract. You pick a concurrency tier, and it comes with a bandwidth allotment. There are two families, matched to the two main workload shapes.

Sticky sessions (Ports)

Sold by Ports, from 1 to 50. Each port holds one concurrent connection, and the assigned IP persists for 10 minutes or more. Use sticky sessions for multi-step, session-bound work: logging in, paginating, filling carts, anything where the IP needs to stay stable across requests.

Tier	Price	Bandwidth	Price per GB	Auth IPs
1 Port	$50/mo	~17 GB	$3.00	4
10 Ports	$400/mo	~167 GB	$2.40	11
50 Ports	$1,750/mo	~833 GB	$2.10	25

High rotation (Threads)

Sold by Threads, from 40 to 650. Every request can exit through a new residential IP. Use high rotation for large-scale, parallel collection where IP diversity matters more than session continuity.

Tier	Price	Bandwidth	Price per GB	Auth IPs
40 Threads	$120/mo	40 GB	$3.00	4
250 Threads	$600/mo	250 GB	$2.40	11
650 Threads	$1,365/mo	650 GB	$2.10	25

These are anchor tiers. The pricing page has the full ladder of seven tiers per family and always shows current rates.

Bandwidth top-ups

If you exceed your plan’s allotment, you buy a top-up, from 10 GB to 1 TB, on the same per-GB curve ($3.00 down to $2.10). You can top up instantly from your members panel with no plan change. Top-ups are one-time purchases billed by usage, and unused gigabytes roll over to the following months rather than expiring.

What is included

Because you are billed for bandwidth, several things other providers list separately are simply included:

• Both authentication methods. IP whitelisting and username/password both work on your ports at no extra cost. The plan’s Auth-IP allowance (4 to 25 addresses by tier) is a whitelist quota, not a per-IP charge.
• All protocols on every plan. HTTPS and SOCKS5, with country and region targeting. Sticky plans cover US, CA, GB, and EU; rotation plans run on an international mix that excludes US, CA, GB, EU, AU, and NZ.

How to size a plan

Two numbers decide your tier: how much data you will move, and how much concurrency you need.

1. Estimate your bandwidth. Multiply your monthly request count by your average response size (sample 50 to 100 real responses and measure the payload in KB, rather than guessing from the HTML source). That gives total gigabytes.
2. Estimate your concurrency. Divide your target throughput by the time each request takes. If a request takes 2 seconds and you need 1 request per second, you need about 2 concurrent connections; add headroom, so 5 to 10 in practice.
3. Pick the tier that covers both. Choose the Ports or Threads tier whose bandwidth allotment covers your data and whose concurrency covers your parallelism. If your data runs ahead of your concurrency, add top-ups rather than jumping to a bigger tier.

Worked example. A price-monitoring job hits 250,000 lightweight JSON endpoints a month at 8 KB each, about 2 GB of data, but needs roughly 10 concurrent workers to finish on time. Concurrency, not data, sets the tier here: a small rotation plan covers the 10 workers, and the 2 GB sits well inside its allotment. A heavy scraping job pulling 400 GB of full HTML pages is the opposite: pick the tier whose bandwidth covers 400 GB, and top up if a busy month runs over.

For a deeper treatment of the concurrency side, see how many IPs you need for web scraping. For the model-versus-model view (bandwidth vs per-IP vs per-request), see proxy bandwidth vs IP count pricing.

Pricing for agent workloads

Bandwidth pricing suits autonomous agents better than flat per-IP plans. Agent traffic is bursty: it sits idle, then spikes. A flat per-IP plan makes you provision for the peak and pay for the idle time, while a bandwidth model charges only for what actually flows, so an idle agent costs nothing and a busy one pays only for the data it moves.

See our AI agent proxy use cases for more.

The bottom line

LocalProxies pricing is deliberately simple: you pay for bandwidth, provisioned through a Ports or Threads tier, at a rate that drops as you scale, with the extras other providers meter separately folded in. To size a plan, estimate your data and concurrency, then match a tier on the pricing page. Or start a free trial and measure your real consumption before you commit.