API pricing A delivered page costs 5 to 20 credits
A plain fetch is 5 credits. JavaScript rendering, residential routing and a screenshot each add 5. Costs add and never multiply, and a request that ends without a result costs nothing. These plans are for the API. Managed scraping is quoted separately.
500 credits after you confirm your email. No card.
Every plan can use every request type
A plan sets your credits, how many requests run at once and how many async jobs can wait in the queue. Everything else is on every plan, the trial included.
Trial
$0one time, no card
For trying every request type on your own pages before you pay.
- 500 credits, one time
- 1 request at a time
- 10 async jobs waiting
Developer
$49.99a month
For up to 10,000 datacenter HTTP requests, or 2,500 residential browser requests with screenshots, each month.
- 50,000 credits a month
- 5 requests at a time
- 250 async jobs waiting
Startup
$129a month
For up to 30,000 datacenter HTTP requests, or 7,500 residential browser requests with screenshots, each month.
- 150,000 credits a month
- 25 requests at a time
- 1,000 async jobs waiting
Business
$289a month
For up to 70,000 datacenter HTTP requests, or 17,500 residential browser requests with screenshots, each month.
- 350,000 credits a month
- 100 requests at a time
- 5,000 async jobs waiting
Credits on a paid plan reset to the plan allowance at each renewal. Unused credits do not carry over.
Taxes may be added at checkout based on your billing location.
On every plan, the trial included
Output
- HTML, visible text or Markdown, chosen per request, all at the same price
- Screenshots of the viewport, the full page or one element
Browser
- JavaScript rendering in a real browser
- Desktop or mobile viewport
- Wait for the page to load, for a CSS selector up to 15 seconds, or a fixed delay up to 30 seconds
- Images, styles, fonts and ad scripts blocked by default
Routing
- Datacenter and residential proxy pools
- Country targeting on the residential pool
- Sticky sessions
- Your own proxy if you prefer
Delivery
- Sync responses up to 120 seconds
- Async jobs up to 130 seconds, polled by job id or posted to your HTTPS webhook with your own reference id
- Custom headers forwarded to the site
- Standard mode, or Auto mode with a maximum credit ceiling
- One active API key and job history in your dashboard
What one request costs
In Standard mode the route you choose sets the price before the request runs. Prices are the same on every plan.
Datacenter HTTP
5 credits
Datacenter browser
10 credits
Residential HTTP
10 credits
Residential browser
15 credits
Any screenshot
+5 credits
A screenshot needs a browser route, so the cheapest screenshot request is 15 credits. The most one request can cost is 20. In Auto mode, max_credits caps what a request can charge.
What a month of requests costs
Multiply the cost of a delivered page by the number of pages delivered.
10,000 datacenter HTTP requests
5 credits each, 50,000 in total. Covered by Developer.
5,000 datacenter browser requests
10 credits each, 50,000 in total. Covered by Developer.
2,000 residential browser requests
15 credits each, 30,000 in total. Covered by Developer.
10,000 residential browser requests with screenshots
20 credits each, 200,000 in total. Covered by Business.
Auto mode may settle on a cheaper route than the one you would have chosen, so a real month can cost less. It never costs more than your ceiling.
What you are charged, and when you are not
Standard mode
The request you send fixes the price. A successful result charges that price once. Recovery attempts inside ScrapeDrive add nothing, and a request that ends without a result settles at zero.
Auto mode
You set max_credits as a hard ceiling. ScrapeDrive charges the configuration that succeeded, once, and never more than your ceiling. Attempts that failed on the way add nothing.
Rejected requests
A request rejected for validation, authentication, balance or account limits costs nothing. A 429 is a rejection.
Completed requests
A request that completes is charged even when the page it returns is not what you expected, including a site error page. Check results for your use case.
How many requests run at once
Concurrency and the async queue are the limits that change with the plan.
| Plan | Requests at once | Async jobs waiting |
|---|---|---|
| Trial | 1 | 10 |
| Developer | 5 | 250 |
| Startup | 25 | 1,000 |
| Business | 100 | 5,000 |
Each plan also has a per-minute safeguard. A request over either limit is rejected with a 429 and costs nothing. Back off and retry, or move to a larger plan.
Running out, changing plans and cancelling
What happens to your credits and your bill in each case.
- When credits run out
- The API rejects the next request and charges nothing. There is no overage billing. Renew early for a fresh allowance, or move to a larger plan.
- Renew early or move up
- You pay the full plan price that day, receive the plan’s full allowance and start a new billing cycle.
- Move down
- A smaller plan starts at your next renewal, with no proration. You can also switch now, paying the smaller plan’s full price for a fresh allowance and a new billing cycle.
- Cancel
- When your subscription ends, API access stops and any remaining credits are forfeited.
API plans or managed scraping
These plans pay for API requests your own code sends. Managed scraping is a separate service: our team builds, runs and maintains the collection, and each project is scoped and quoted on its own, outside these plans.
Questions about pricing
Start with 500 credits
500 credits after you confirm your email. No card.