Wikidata Entity Batch Export

What can this tool do?

Resolve a supplied list of Wikidata Q/P identifiers into multilingual labels, descriptions, aliases and selected Wikipedia sitelinks. Useful for turning opaque IDs in a knowledge graph or research export into readable records.

How to use it

{"entityIds":["Q148","Q1490"],"languages":["en","zh"],"maxResults":20,"maxRequests":3}

Accepts up to 100 supplied identifiers and five language codes. IDs are deduplicated before requests. It does not discover companies, execute SPARQL, export claims, download media or verify Wikidata's factual accuracy.

Each found row contains its identifier, language-qualified metadata, available revision ID, source URL and CC0 notice. Missing entities are explicit notFound rows; missing labels remain missing rather than invented.

Pricing

$0.50 per 1,000 delivered found entities ($0.0005 each), plus $0.00005 when a run starts at the supported 256–512 MB memory range. Missing-entity status rows have no entity fee. The start fee can still apply to an unsuccessful run. Set a spending limit before running.

Is it legal?

Uses the official MediaWiki Action API with maxlag and stops on throttling. Structured Wikidata data is CC0; other Wikimedia content has separate terms. This tool does not download linked pages or images and is not endorsed by Wikimedia.

Source guide: https://www.wikidata.org/wiki/Wikidata:Data_access

What data does it return?

Results are written to the Apify dataset. OUTPUT reports whether the export completed and why it stopped. RESULTS_CSV contains the rows delivered in this run. CHECKPOINT can continue with the same source inputs; omit it for a fresh export.

Requests are sequential within a run, restricted to the documented source, and bounded by timeout, response size, request count and output count. There are no automatic retries, proxy rotations or redirect fallbacks. Source failures are reported separately from an empty or absent result.

Continuation is best effort. A crash between delivery and checkpoint persistence can repeat rows; the source can change between runs. This is not an immutable snapshot, exactly-once delivery or monitoring. CSV text that could be interpreted as a spreadsheet formula is prefixed with an apostrophe.

FAQ

Can I use the original source for free? Yes. The Actor fee pays for the packaged batch workflow and Apify delivery, not exclusive access to the public source.

Does a completed export prove the source is complete or correct? No. Read the stated coverage, status, source dates and limitations before relying on a row.

Feedback

Use the Actor's Issues tab for a reproducible problem. Include a small public example, OUTPUT and the expected result. Do not post tokens, confidential inventories or customer data in a public issue. Report upstream data errors to the original provider.