Site coverage: modes, fields & pricing

Which Asian sites have a working public scraper, what each one can do, and exactly which columns you get back. Every cell is read out of the actor's own schema — no feature blurbs, no roadmap items.

Snapshot 2026-09-07 · 40 sites · 86 modes · 642 documented fields · 15 with both 繁中 and English listings

How to read this

Modes are the unit of work, not the actor

Most actors do more than one job. A forum actor usually separates thread-list (cheap, wide) from thread (expensive, deep) so that sweeping a board and reading one discussion are not billed at the same rate. Pick the mode before you pick the actor — it is what determines both the fields you get and what you pay.

Field names are the contract

Output fields are snake_case English on every actor regardless of the site's language, a missing value comes back as null rather than being dropped, and every record carries scraped_at as an ISO 8601 timestamp. That is what makes two sites' output joinable in the same dashboard.

A range in the price column means several event types

Pay-per-event pricing bills the work, not the run. Listing rows are the cheap end; detail pages and individual comments cost more because they cost more to fetch. Apify adds a $0.005 start fee once per run, which is not included above — across a thousand items it disappears, across ten it is your largest single cost.

What is deliberately not here

Sites where the only route in is a headless browser. At current pricing a browser-bound scraper runs at a negative margin, so it does not get built — the reasoning, with measured per-item costs, is in the forum scrapability table.