# Sila vs Scrape.do

Scrape.do is a proxy-based scraping API sold by request volume and concurrency, with a markdown output option. Sila is a markdown-first scrape and crawl API with page signals, sold by credits. They overlap when you want a page fetched and cleaned; they differ in how you pay and what comes back.

Facts checked 2026-11-05

## Choose Sila when

-   You want clean markdown plus structured page signals as the default result, not raw HTML you convert yourself.
-   You want to crawl a whole site from a sitemap or llms-full.txt, with status, pages and webhooks, instead of issuing one request per URL.
-   You want a two-number credit rule (1 static, 5 headless) rather than a cost that varies by proxy type and render mode.
-   You want MCP, a CLI and agent skills to drive it from an AI agent.

## Choose Scrape.do when

-   You need geo-targeted or residential proxies and fine control over proxy type; Sila has no proxy-type options.
-   You need high request volume or concurrency at a low per-request cost: Scrape.do's plans run from 250,000 to 10,000,000 credits and 10 to 200 concurrent requests.

## Facts side by side

| Attribute | Sila | Scrape.do |
| --- | --- | --- |
| Free allowance | 500 credits/month | 1,000 successful credits, 5 concurrent |
| Paid plans | Pro $19/mo (20,000 credits); Scale $99/mo (200,000) | Hobby $29/mo (250,000 credits, 10 concurrent); Pro $99/mo (1,250,000, 50); Business $249/mo (3,500,000, 100); Advanced $699/mo (10,000,000, 200) |
| Billing rule | 1 credit per statically fetched page, 5 credits per page that needed the headless browser or a screenshot | Only successful requests are charged; the cost depends on proxy type and render mode |
| Markdown output | Default format | Available with output=markdown |
| Page signals in the response | Yes: title, description, language, category, JSON-LD, forms and detected integrations | Not listed on the pages checked |
| Whole-site crawl | Crawl API with sitemap, llms-full.txt and incremental re-crawls | Request-based scraping API |
| Agent tooling | Hosted MCP server (OAuth or API key), CLI (npx @sila-sh/cli) and installable agent skills | Not listed on the pages checked |

Free allowance

Sila500 credits/month

Scrape.do1,000 successful credits, 5 concurrent

Paid plans

SilaPro $19/mo (20,000 credits); Scale $99/mo (200,000)

Scrape.doHobby $29/mo (250,000 credits, 10 concurrent); Pro $99/mo (1,250,000, 50); Business $249/mo (3,500,000, 100); Advanced $699/mo (10,000,000, 200)

Billing rule

Sila1 credit per statically fetched page, 5 credits per page that needed the headless browser or a screenshot

Scrape.doOnly successful requests are charged; the cost depends on proxy type and render mode

Markdown output

SilaDefault format

Scrape.doAvailable with output=markdown

Page signals in the response

SilaYes: title, description, language, category, JSON-LD, forms and detected integrations

Scrape.doNot listed on the pages checked

Whole-site crawl

SilaCrawl API with sitemap, llms-full.txt and incremental re-crawls

Scrape.doRequest-based scraping API

Agent tooling

SilaHosted MCP server (OAuth or API key), CLI (npx @sila-sh/cli) and installable agent skills

Scrape.doNot listed on the pages checked

## Sources, checked 2026-11-05

-   [scrape.do/pricing](https://scrape.do/pricing)
-   [scrape.do/documentation/request-costs](https://scrape.do/documentation/request-costs)

## 500 free credits every month.

Create a key in under a minute and scrape your first page.

[Get API key](https://sila.sh/login)