v1.13.1-4059c5c7
Skip to main content
ProxiesGuide

Proxy Pricing Models Explained: Per-GB vs Per-IP vs Per-Request

9 min read

By Hex Proxies Engineering Team

Proxy pricing is one of the most confusing aspects of choosing a provider. Some charge per gigabyte of data, others per IP address, and some per request or API call. The right model for you depends entirely on your use case, volume, and the type of data you are collecting.

Choosing the wrong pricing model can cost you 3-5x more than necessary. A scraper that downloads lightweight text data wastes money on a per-IP plan with unused bandwidth. A multi-account manager running 100 accounts wastes money on a per-GB plan when each account generates minimal traffic.

This guide breaks down each pricing model with concrete examples, cost calculations, and recommendations so you can make an informed decision.


The Three Main Pricing Models

1. Per-GB (Bandwidth-Based)

How it works: You pay for the amount of data transferred through the proxy, measured in gigabytes. You can use as many IPs as the provider's pool offers, send as many requests as you want, and the only meter running is your bandwidth consumption.

Typical pricing range: $2-15/GB depending on proxy type and provider.

Hex Proxies residential pricing: Starting at $4.25/GB with volume discounts available for higher tiers.

Best for:


  • Web scraping (especially text-heavy pages)

  • SERP monitoring

  • Price comparison data collection

  • Any high-request-count, low-bandwidth task


Example cost calculation:

ScenarioPagesAvg Page SizeTotal DataCost at $4.50/GB
Product scraping (text only)10,000150 KB1.5 GB$6.75
Full page scraping (with images)10,0002 MB20 GB$90.00
API data collection100,0005 KB0.5 GB$2.25
SERP monitoring5,000200 KB1 GB$4.50
Pro tip: Block images, CSS, and JavaScript files you do not need. A page that transfers 2 MB with all resources drops to 100-300 KB when you only fetch the HTML. This can reduce your costs by 80%+.

Use our proxy cost calculator to estimate costs for your specific workload.

2. Per-IP (Address-Based)

How it works: You pay a monthly fee for each IP address assigned to you. The IP is yours for the duration of the billing period, and bandwidth is typically unlimited. You get a fixed number of static IPs that do not change.

Typical pricing range: $1.50-5/IP/month for ISP proxies, $0.50-3/IP/month for datacenter proxies.

Best for:


  • Account management (one IP per account)

  • Sneaker botting and checkout automation

  • Long-running sessions requiring consistent identity

  • Tasks where unlimited bandwidth on a fixed IP is more valuable than IP diversity


Example cost calculation:

ScenarioIPs NeededCost at $3/IP/moMonthly Cost
10 social media accounts10$3.00$30
50 marketplace accounts50$3.00$150
5 sneaker bot tasks5$3.00$15
200 account management200$3.00$600
Per-IP pricing is predictable: you know your exact cost before the month starts. There are no surprise bandwidth bills. However, you get no IP rotation -- if an IP gets banned, you need to request a replacement or purchase additional IPs.

Learn more about ISP proxy pricing and features on our ISP proxies page.

3. Per-Request (API-Based)

How it works: You pay a fixed price for each HTTP request routed through the proxy. The provider handles IP rotation, browser rendering, CAPTCHA solving, and retry logic. You send a URL, and get back the rendered content.

Typical pricing range: $0.50-5 per 1,000 requests (varies wildly by provider and target site difficulty).

Best for:


  • Teams that want a managed scraping solution

  • High-difficulty targets that require browser rendering

  • Low-volume, high-value data extraction

  • Companies without in-house scraping expertise


Example cost calculation:

ScenarioRequestsCost at $1/1,000Monthly Cost
Daily price checks (100 products)3,000$1.00$3
Weekly SERP monitoring (500 keywords)2,000$1.00$2
Large-scale crawl (1M pages)1,000,000$1.00$1,000
E-commerce monitoring (50K products daily)1,500,000$1.00$1,500
Per-request pricing looks cheap at small scale but becomes expensive fast. At high volumes, per-GB pricing almost always wins because you are paying for raw bandwidth rather than a managed service markup.

Head-to-Head Comparison

FactorPer-GBPer-IPPer-Request
Cost predictabilityMedium (depends on page size)High (fixed monthly)Medium (depends on volume)
Best volume levelMedium to highLow to mediumLow to medium
IP rotationAutomatic, unlimitedNone (static IPs)Managed by provider
Bandwidth includedMeteredUnlimitedN/A (per request)
Scaling cost curveLinearLinearLinear but expensive
Infrastructure neededYour own scraperYour own toolsMinimal (API calls)
FlexibilityMaximumGood for specific use casesLimited to provider's API

How to Choose: Decision Framework

Step 1: Determine your primary use case

  • Scraping public data at scale -> Per-GB
  • Managing persistent accounts -> Per-IP
  • Quick data extraction, no infrastructure -> Per-Request

Step 2: Estimate your volume

Calculate your expected monthly usage in all three dimensions:

  • Bandwidth: Number of pages x average page size (in MB) = total GB
  • IPs needed: Number of persistent identities needed simultaneously
  • Request count: Total HTTP requests per month

Step 3: Calculate costs under each model

Use the tables above or our cost-per-request calculator to compare.

Step 4: Consider hidden costs

Per-GB hidden costs:


  • Wasted bandwidth from retries and blocked requests

  • No control over which IP you get (rotation is random)


Per-IP hidden costs:

  • Replacement fees if IPs get banned

  • Paying for IPs during downtime or off-peak periods

  • Need to purchase more IPs to replace blocked ones


Per-Request hidden costs:

  • Premium charges for difficult targets

  • Rate limits that throttle your speed

  • Vendor lock-in (your code depends on their API format)



Real-World Scenarios: Which Model Wins?

Scenario 1: Price monitoring for an e-commerce company

Task: Scrape 50,000 product pages daily, extracting price, title, and availability (text only).

  • Average page: 150 KB of HTML (blocking images)
  • Monthly data: 50,000 x 150 KB x 30 days = ~225 GB
Cost comparison:
ModelCalculationMonthly Cost
Per-GB ($4.50/GB)225 GB x $4.50$1,012
Per-Request ($1/1K)1.5M requests x $1$1,500
Per-IP (N/A)Would need 1000+ rotating IPsNot applicable
Winner: Per-GB -- saves $488/month over per-request, and per-IP is not suitable for rotating scraping.

Scenario 2: Managing 200 social media accounts

Task: Each account posts once daily and checks messages. Minimal bandwidth per account.

  • Bandwidth per account: ~50 MB/month
  • Total bandwidth: 200 x 50 MB = 10 GB
  • IPs needed: 200 (one per account, persistent)
Cost comparison:
ModelCalculationMonthly Cost
Per-IP ($3/IP)200 IPs x $3$600
Per-GB ($4.50/GB)10 GB x $4.50$45
Wait -- per-GB looks cheaper here. But there is a catch: per-GB residential proxies rotate IPs. You cannot guarantee the same IP for each account. You would need sticky sessions, which have timeout limits. For persistent identity, per-IP (ISP) is the correct choice despite the higher cost.

Winner: Per-IP -- the use case requires IP persistence, which per-GB rotation cannot reliably provide.

Scenario 3: Occasional competitive research

Task: A marketing team checks competitor pricing weekly, scraping 500 pages each Monday.

  • Monthly requests: 500 x 4 = 2,000
  • Bandwidth: 2,000 x 200 KB = 400 MB
Cost comparison:
ModelCalculationMonthly Cost
Per-GB ($4.50/GB)0.4 GB x $4.50$1.80
Per-Request ($1/1K)2K requests x $1$2.00
At this volume, the cost difference is negligible. Choose based on convenience: per-GB if you have scraping infrastructure, per-request if you want a simple API call.

Winner: Tie -- both are under $2/month at this volume.


Hex Proxies Pricing at a Glance

ProductModelStarting PriceKey Features
Residential ProxiesPer-GB$4.25/GB10M+ IPs, 93+ countries, auto-rotation, sticky sessions
ISP ProxiesPer-IPPer-IP/month250K+ IPs, unlimited bandwidth, static dedicated IPs
Both products include:
  • Full dashboard with usage analytics
  • API access for programmatic management
  • No minimum commitment
  • 24/7 support

Common Pricing Mistakes to Avoid

1. Not blocking unnecessary resources

If you are scraping product data, you do not need images, CSS, or JavaScript. Loading these resources inflates your bandwidth bill by 5-10x on per-GB plans.

2. Choosing per-request for high volume

Per-request pricing includes a service margin on top of the raw proxy cost. At 100,000+ requests per month, per-GB is almost always cheaper because you are paying for infrastructure, not a managed service.

3. Over-provisioning IP addresses

On per-IP plans, only purchase the IPs you actively use. Ten idle IPs at $3/month is $30 wasted. Right-size your IP pool and scale up only when needed.

4. Ignoring bandwidth optimization

Simple techniques like gzip compression, HTML-only fetching, and response streaming can reduce bandwidth by 60-80%, directly cutting your per-GB costs.

5. Not accounting for retries

If your success rate is 80%, one in five requests is wasted bandwidth. Improving your scraping strategy (better headers, appropriate delays, right proxy type) reduces retries and saves money.


Calculate Your Costs

Use our free calculators to estimate your proxy costs:


Choose the Right Plan for Your Use Case

Now that you understand the three pricing models, you can make an informed choice. Most scraping teams start with per-GB residential proxies and add per-IP ISP proxies for account management workflows.

View all Hex Proxies plans -- transparent pricing, no hidden fees, and volume discounts available for all plan types.