Apify

platform

proprietary(Apify-Cloud,Actor-store,managed)

source

partial(Crawlee SDK published, the platform that runs Actors is not)

oss-sdk

Crawlee(Apache-2.0,OSI, JS+Python)

license

mixed(platform closed, SDK open)

The product proper - the Apify platform and its Actor store - is proprietary SaaS. Apify also maintains Crawlee, an Apache-2.0 scraping library, but Crawlee is an adjacent SDK rather than the platform core, so most of the value stays behind the managed cloud. Open core would mean an OSI-licensed core with functionality withheld for a paid tier; there is no open core here, only an open periphery. Publishing a repository while the runtime itself does not ship counts as partial source, which scores 2, source-available.

• https://apify.com recorded 2026-08-13

  Proprietary full-stack platform - Apify Store, Actors ("Build and run serverless programs"), Integrations, an MCP server, Anti-blocking and Proxy - with a single "Open source" navigation heading holding Crawlee alone. The store now advertises 59,817 Actors, up from the 35,000+ this record carried. No self-hosted Actor runtime is offered.

• https://github.com/apify/crawlee recorded 2026-08-13

  Crawlee repo - Apache-2.0 license badge and repo metadata spdxId Apache-2.0, "Apache License 2.0"; LICENSE.md at the root. The library only; it is not the Actor platform.

Crawlee at 25,373 GitHub stars (corroborating, not basis); the platform serves named enterprises (T-Mobile, Microsoft, Accenture) and the store advertises 59,817 published Actors, which implies a sizable active developer base. The site links a Discord invite without stating a member count, so no community figure is carried. No precise account, run or API-call count is published, so the level stays an aggregate judgment across platform + OSS SDK and remains 3.

• https://apify.com recorded 2026-08-13

  "Browse 59,817 Actors" in the store; T-Mobile and Accenture among named enterprise users; a Discord invite link. NOT on the page: a Discord member count, an account count, a run count or an API-call figure.

• https://github.com/apify/crawlee recorded 2026-08-13

  25,373 stargazers on the OSS Crawlee library. Corroborating attention signal, not the basis.

Broad scrape/browse feature matrix, multi-engine crawling, proxy/anti-block, structured extraction, serverless deploy, and MCP integration for agents. Among the most complete scrape platforms; rated 4 (frontier-adjacent for the browser/scrape sub-segment).

• https://apify.com recorded 2026-08-13

  "Browse 59,817 Actors"; navigation items for Actors, Integrations, MCP ("Give your AI access to Actors"), Anti-blocking ("Scrape without getting blocked") and Proxy ("Rotate scraper IP addresses"); "Configure your Apify MCP server with Actors and tools for seamless integration with MCP clients".

• https://github.com/apify/crawlee recorded 2026-08-13

  "Works with Puppeteer, Playwright, Cheerio, JSDOM, and raw HTTP. Both headful and headless mode. With proxy rotation." Feature list adds structured storage of tabular data and files, automatic scaling, "Integrated proxy rotation and session management", hooks and a bootstrap CLI. NOT named: Selenium.