AI Data Extractor & AI Web Scraper: Pages to Structured JSON
Pricing
from $3.50 / 1,000 page extracteds
Turn web pages, or another Actor's dataset, into structured JSON with an LLM: an AI web scraper for pages you already have. Give a JSON schema or plain-English fields; every answer is validated, and you pay $5 per 1,000 valid pages plus the model's tokens. No browser.
Versions follow MAJOR.MINOR.PATCH (src/version.py); Apify shows MAJOR.MINOR from .actor/actor.json.
Every run logs its version and records it in the RUN_STATS key-value record.
About future changes in output: what a model reads off a page can differ between models, and between runs of the same model. Every answer is checked against your schema, and one that doesn't match comes back with
valid: falseand the reasons, uncharged.
RUN_STATS keeps the reason as before.valid: false, uncharged.mms_llm) used by this actor and by AI for Every Row, which runs the same validated call on dataset rows
instead of web pages.null, blank or "N/A"-style, or only empty lists) now comes back valid: false with the reason,
uncharged (pagesEmpty in RUN_STATS). One real value, false and 0 included, still makes a valid page.botChecks), and thin pages that a
script fills in (under 15 words outside the menus and footer plus scripts, or the data waiting in an attribute
for a script, with no JSON-LD to read instead) are reported as such, not sent. Catches the three pages of the
2026-10-01 test that 1.0.2 charged for nothing; every page it read correctly is still read.fullPage
setting, text inside a <template shadowrootmode> goes to the model too, as a browser would show it.First release: web pages, or the pages in another actor's dataset, to JSON that matches your schema.
urls, or datasetId (+ optional datasetUrlField) to read page URLs from any dataset and chain
after another actor with {{resource.defaultDatasetId}} (the shared reader of page-to-markdown and seven more
actors: url, pageUrl, link, website, loadedUrl found automatically; Google Maps links never used;
tracking parameters dropped; each row carries sourceTitle, sourcePlaceId, sourceIndex).fields in plain English (each becomes a camelCase key; a hint after a colon goes to the
model) or a JSON Schema in schema (overrides fields), plus optional instructions.pageContent fullPage (default) keeps headers, footers and sidebars, minus scripts, forms, hidden elements and
navigation menus; mainContent sends only the main content. The page's schema.org JSON-LD goes along.
maxTokensPerPage (default 8,000) is a hard cap: a longer page keeps its start and end, truncated: true.model takes any OpenRouter id (default openai/gpt-4.1-mini). The schema
goes as a structured-output format, strict when it qualifies; a provider that refuses it in strict mode gets it
again as a hint. Every answer is validated here: valid, errors (at most 5), data kept for inspection.
A proxy that refuses the run (401/402/403) stops it at once, with the reason.page-extracted event is charged once per page whose answer validated,
after the row is in the dataset. Failed, blocked and invalid pages are rows with the reason, never charged.
maxPages and the maximum cost per run are honoured before any page is fetched: a slot is claimed per page and
given back when the page doesn't validate.