Pornhub Model Extractor| Videos & Profile

Pornhub Model Extractor

Extract profile information (INFO) or video-listing metadata (VIDEOS) from model and pornstar profile pages. This tool is for adults and lawful, authorized use only. Respect applicable law, privacy, site terms and access restrictions. It does not bypass login, age verification, geographic blocks or other access controls.

Inputs

• startUrls: HTTPS model or pornstar URLs on www.pornhub.com, with or without /videos. PPE rejects credentials, other domains and malformed profile paths.
• type: INFO or VIDEOS (default).
• startPage: initial video-listing page; 0 means the initial page. It does not apply to INFO.
• endPage: stop at this listing page; 0 means no page bound. PPE follows only explicit next links for the same profile, not guessed URLs.
• maxItems: output limit. On Free/non-paying usage the existing limit is 50; PPE and other paying usage use the supplied value or the runtime default of 5. The UI supplies 50 by default.
• maxRequestsPerCrawl: crawler request bound; the UI defaults to 50 and the runtime fallback is 100.
• proxyConfiguration.useApifyProxy: optionally use Apify Proxy where access is permitted. A proxy does not guarantee availability or authorize bypassing restrictions.
• nameDataset: legacy setting; it does not redirect output to a named dataset.
• useRandomDelays: retained for compatibility but is not wired to request delays.

Provide explicit inputs before running: omitted input can activate the bundled default profile. maxItems counts emitted records, not attempted pages.

Output and PPE validation

INFO records include model name, bio/about, profile attributes, rank, views, subscribers, social links and verification flags where present. A billable profile requires a name in a recognized profile heading plus actual bio/about, a nonempty attribute or an explicitly extracted numeric counter. An explicit zero counter qualifies; an absent field defaulting to zero does not.

VIDEOS records include title, canonical video URL/viewkey, video ID and available duration, views, rating, thumbnail and preview. PPE requires a nonempty title and valid canonical video URL or viewkey. Duration is optional; absent or invalid duration is represented as 0. PPE deduplicates canonical video identities, including links differing only in tracking parameters or fragments. Missing optional data is not independently proof of a valid result.

PPE rejects recognized error/access-gate pages, non-success HTTP responses, cross-profile redirects and mismatches between requested and final page identity. Gate or login markup hidden in the DOM can conservatively reject an otherwise usable page. Detection is not universal: HTML changes and unrecognized gates can still prevent or affect extraction. Free parsing and pagination retain their legacy behavior, including legacy parsing limitations.

The result writer serializes output, deduplicates results and respects output/event capacity. Saving a record and charging are separate operations, not an atomic transaction. An uncertain save or charge halts further output and requires reconciliation before resuming; it is not silently replayed.

Pricing

The current Apify Pricing tab is authoritative. Free/pay-per-usage execution may still incur platform usage costs for the person running the Actor. In pay-per-event (PPE), the customer pays configured events and the developer pays platform usage, including proxy costs.

PPE uses the result-item event for each accepted, unique output item. Do not configure apify-default-dataset-item alongside it. A configured positive result-item price is required. If Actor Start is enabled, its US$0.00005 per start per GB (minimum 1 GB) may be charged even if no result is produced. This is not a promise of zero cost for an empty run, nor a fixed quote for the result event. Check Pricing before running.

Development

Use Node 20.20.2 and npm 10.8.2:

npm ci --workspaces --include=dev --ignore-scripts --audit=false --fund=false
npm run build

Dependencies and the Docker base image are pinned. The Docker build includes a network-isolated diagnostic. start:prod starts the production crawler; it is separate from compilation.

Limitations

Synthetic tests do not establish live-site compatibility, CDN or proxy availability, or financial settlement. Site HTML and access requirements can change. Confirm applicable site permissions and current pricing before use.