# Scrapier — TikTok API > Scrapier endpoints for TikTok: TikTok Ad Library Leads, TikTok Ad Library, TikTok Ads Top, TikTok Collection Videos and 24 more. Base URL: https://api.scrapier.io/api/v2 · Auth: header `X-API-Key: ` · Synchronous: HTTP 200 returns {"success": true, "data": {...}} · Paginate by sending data.next_cursor back as `cursor` until it is null · Paths keep the trailing slash. Live parameter source (no auth): https://api.scrapier.io/api/v2/endpoints · Docs: https://docs.scrapier.io/api-reference/tiktok.md ## GET /tiktok-ad-library-leads/ TikTok Ad Library Leads: Structured TikTok data: every input is optional. Single lookup, 3 credits per successful call. Cost: 3 credits per successful call · Pagination: single lookup · Results: data.leads Parameters: - keyword (optional, string, default Skincare): Optional search term for the default single-query mode. Single lookup -- next_cursor is always null. - country (optional, string, default US): (no description yet) - period (optional, integer, one of: 7|30|180, default 180): (no description yet) - order_by (optional, string, one of: for_you|impression|ctr|play_2s_rate|play_6s_rate|cvr|like, default for_you): (no description yet) - ad_format (optional, string, one of: All ad types|Spark ads|Non-Spark ads, default All ad types): (no description yet) - ad_language (optional, string, default en): (no description yet) - like_tier (optional, integer): (no description yet) - industry (optional, string, default Skincare): (no description yet) - queries (optional, string): Optional JSON array of per-search override objects to sweep many keyword/industry/country combinations in ONE call. - max_items (optional, integer, default 30): Ads collected per query before dedup/grouping. - max_landing_pages_per_advertiser (optional, integer, default 3): (no description yet) - max_concurrent_queries (optional, integer, default 2): (no description yet) - max_concurrent_details (optional, integer, default 2): (no description yet) - max_concurrent_crawls (optional, integer, default 10): (no description yet) - max_retries (optional, integer, default 5): (no description yet) - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. No parameter is marked required; send at least one input (see descriptions). Example: curl "https://api.scrapier.io/api/v2/tiktok-ad-library-leads/" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-ad-library-leads.md ## GET /tiktok-ads/ TikTok Ad Library: Structured TikTok data: send `urls`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.ads Parameters: - urls (required, string, example https://library.tiktok.com/ads?region=all&adv_name=NVIDIA%20GmbH&adv_biz_ids=6891503886842987266&query_type=2&sort_type=last_shown_date,desc): TikTok Ad Library search URL (filters: region, start_time/end_time, adv_name, adv_biz_ids, query_type, sort_type, ad_type, ad_status, ages, ad_reach, gender) or a bare keyword. One page = 12 ads (TikTok's fixed page size). Required on first call; ignored when `cursor` is supplied (filters are locked into it). - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - include_details (optional, boolean, default true): Fetch each ad's detail record (paid-by, per-region impressions, targeting). Best-effort: if TikTok rate-limits it, ads come from search-card data (detail_fetched=false) with a `warning`. Locked into the cursor. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-ads/?urls=https%3A%2F%2Flibrary.tiktok.com%2Fads%3Fregion%3Dall%26adv_name%3DNVIDIA%2520GmbH%26adv_biz_ids%3D6891503886842987266%26query_type%3D2%26sort_type%3Dlast_shown_date%2Cdesc" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-ads.md ## GET /tiktok-ads-top/ TikTok Ads Top: Structured TikTok data: every input is optional. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.items Parameters: - keyword (optional, string): Optional search term (login-gated by TikTok: echoed, not applied). Each call returns one leaderboard partition (<=20 ads); pass next_cursor to walk deeper into the same filter combination. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - country (optional, string, default US): ISO-2 country code, validated live against TikTok's top_ads country list. - period (optional, integer, one of: 7|30|180, default 180): Leaderboard lookback window in days. - order_by (optional, string, one of: for_you|impression|ctr|play_2s_rate|play_6s_rate|cvr|like, default for_you): Leaderboard ranking axis. - ad_format (optional, string, one of: All ad types|Spark ads|Non-Spark ads, default All ad types): Ad format filter. - ad_language (optional, string, default en): Ad language code. - like_tier (optional, integer): Engagement-tier bucket (1 = top tier). - industry (optional, string, default Skincare): Industry label, matched live. - max_items (optional, integer, default 20): Ads per page (values >20 clamp to 20 = one TikTok leaderboard partition). Follow next_cursor for more; the pool is a few hundred ads for broad filters. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. No parameter is marked required; send at least one input (see descriptions). Example: curl "https://api.scrapier.io/api/v2/tiktok-ads-top/" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-ads-top.md ## GET /tiktok-collection-videos/ TikTok Collection Videos: Structured TikTok data: send `urls`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.videos Parameters: - urls (required, string, example https://www.tiktok.com/@tiktok/collection/Summer-of-Sports-7394627756635573022): TikTok collection URL (https://www.tiktok.com/@user/collection/Name-), bare collection id, or a vm.tiktok.com share link. One native page (up to 30 videos, under `videos`) per call, paged through TikTok's own offset. Required first call; ignored when `cursor` is supplied. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - include_collection_info (optional, boolean, default true, locked in cursor): Fetch owner profile + collection name / total / cover on page 1 (carried to later pages in the cursor). - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-collection-videos/?urls=https%3A%2F%2Fwww.tiktok.com%2F%40tiktok%2Fcollection%2FSummer-of-Sports-7394627756635573022" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-collection-videos.md ## GET /tiktok-comment-replies/ TikTok Comment Replies: Structured TikTok data: send `urls`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.comments · Long-running (client timeout 210 s or more) Parameters: - urls (required, string, example https://www.tiktok.com/@gordonramsayofficial/video/7229884545150061851): TikTok video URL (/video/ or /photo/), vm./vt./tiktok.com/t/ short link or bare 19-digit video id. One video per call; required on the first call, ignored when `cursor` is supplied. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - comments_per_post (optional, integer, default 100, locked in cursor): Top-level comments to walk over the whole cursor chain, most relevant first (default 100; 0 = all). - max_replies_per_comment (optional, integer, default 1, locked in cursor): Replies per top-level comment (default 1; 0 = the whole thread — long threads are then split across calls by their native reply cursor). - include_top_comments (optional, boolean, default true, locked in cursor): Emit the top-level comment row before its replies. - hosts (optional, string, one of: us|www, default us, locked in cursor): Datacenter walked by this cursor chain: us.tiktok.com (default) or www.tiktok.com — they order top-level comments differently, so each cursor chain walks ONE host. - trim (optional, boolean, default false, locked in cursor): true drops the heavy `native` object and detailed_mentions from every row. - label (optional, string, locked in cursor): Comma list of judge presets to add (sentiment, question, purchase_intent, complaint run by default; spam, toxic, low_quality, injection on request; "none" = off). - judgments (optional, string, one of: on|off, default on, locked in cursor): off = no computed.labels. - drop (optional, string, locked in cursor): Comma subset of spam,low_quality dropped at p >= 0.8 (preset must be in `label`). - evidence (optional, boolean, default false, locked in cursor): 1 adds computed.labels_evidence for the presets named in `label`. - estimate (optional, boolean, default false): 1 = one `type: estimate` row, no fetch, next_cursor null. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-comment-replies/?urls=https%3A%2F%2Fwww.tiktok.com%2F%40gordonramsayofficial%2Fvideo%2F7229884545150061851" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-comment-replies.md ## GET /tiktok-comments/ TikTok Comments: Structured TikTok data: every input is optional. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.comments Parameters: - usernames (optional, string, default mrbeast): TikTok username (@ optional). Required on first call when `manual_video_urls` is not supplied; ignored when `cursor` is supplied. - manual_video_urls (optional, string): TikTok video: full URL, vm.tiktok.com / vt.tiktok.com short link, or numeric video id. Takes precedence over `usernames`. Ignored when `cursor` is supplied. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - results_per_profile (optional, integer, default 30): Max videos discovered per profile. Locked into the cursor. - comments_per_post (optional, integer, default 100): Max top-level comments per video. Video input: empty = walk all comments. Usernames input: default 100 per video. Locked into the cursor. - max_replies_per_comment (optional, integer, default 2): Max replies fetched per comment (capped at 500; 0 = skip reply threads, fastest). Reply threads are fetched in parallel within a ~5s window per call; threads not fetched in time keep reply_comment_total and are listed in `warning`. Locked into the cursor. - profile_sections (optional, string, default videos): Profile section: videos or reposts (first one is used). Locked into the cursor. - profile_sorting (optional, string, default latest): Sort: latest, oldest, or popular. Locked into the cursor. - oldest_post_date (optional, string): YYYY-MM-DD or days ago (e.g. 30). Locked into the cursor. - newest_post_date (optional, string): YYYY-MM-DD. Locked into the cursor. - exclude_pinned (optional, boolean, default false): Exclude pinned posts when scraping profiles. Locked into the cursor. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. No parameter is marked required; send at least one input (see descriptions). Example: curl "https://api.scrapier.io/api/v2/tiktok-comments/" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-comments.md ## GET /tiktok-get-song-details/ TikTok Get Song Details: Structured TikTok data: send `music_ids`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.items · Long-running (client timeout 210 s or more) Parameters: - music_ids (required, string, example 7439295283975702544): TikTok sound / song clip id (7439295283975702544 = SZA "luther"), a music page URL, a video URL (-> the sound it uses) or a vm./vt. short link. Required on the first call; ignored when `cursor` is supplied. One native TikTok item_list page (up to 30 videos) per call in mode videos; every row carries the sound's music_info block. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - mode (optional, string, one of: videos|song, default videos, locked in cursor): videos = one 30-video page per call (native offset cursor); song = single lookup of the sound's own record (one row, next_cursor null). - include_song_details (optional, boolean, default true, locked in cursor): Attach music_info to every row (re-read from /api/music/detail/ on each page). Forced on in mode song. - hydrate_song (optional, boolean, default true, locked in cursor): Video-page hydration of music_info (full credit line, precise durations, matched-song id). - include_artist_profile (optional, boolean, default true, locked in cursor): Artist / owner profile pages -> artists[].is_verified + follower counts. - region (optional, string, locked in cursor): ISO-2 market sent to the music API (also pins the proxy exit country; default US). - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-get-song-details/?music_ids=7439295283975702544" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-get-song-details.md ## GET /tiktok-hashtag-videos/ TikTok Hashtag Videos: Structured TikTok data: send `hashtags`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.videos Parameters: - hashtags (required, string, example travel): Hashtag, with or without # (a tiktok.com/tag/ URL also works). Required on first call; ignored when `cursor` is supplied. One call = one TikTok page (up to 30 videos). - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-hashtag-videos/?hashtags=travel" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-hashtag-videos.md ## GET /tiktok-keyword-search/ TikTok Keyword Search: Structured TikTok data: send `queries`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.videos Parameters: - queries (required, string, example funny cats): Search term. Required first call. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - date_posted (optional, string, locked in cursor): today | this-week | this-month | this-year. - sort_by (optional, string, default relevance, locked in cursor): relevance | date | views. - region (optional, string, locked in cursor): 2-letter country code sent as TikTok search region (default US). - fill_from_hashtag (optional, boolean, default true, locked in cursor): After TikTok's logged-out keyword search ends (~90-130 videos), continue with the #keyword hashtag feed (rows source=hashtag_feed). - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-keyword-search/?queries=funny+cats" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-keyword-search.md ## GET /tiktok-live-info/ TikTok Live Info: Structured TikTok data: send `room_ids`. Single lookup, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: single lookup · Results: data.room Parameters: - room_ids (required, string, example 7523685855395842871): TikTok LIVE room_id. Lookup endpoint -- next_cursor null. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - user_id (optional, string, locked in cursor): Owner cross-check. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-live-info/?room_ids=7523685855395842871" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-live-info.md ## GET /tiktok-live-search/ TikTok LIVE Search: Structured TikTok data: send `keywords`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.live_streams Parameters: - keywords (required, string, example gaming): Live-stream search keyword. Required first call; ignored when `cursor` is supplied. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-live-search/?keywords=gaming" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-live-search.md ## GET /tiktok-playlist-videos/ TikTok Playlist Videos: Structured TikTok data: every input is optional. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.playlists Parameters: - playlist_ids (optional, string, default 7090570072548461358): TikTok playlist id or playlist URL -> playlist mode: ONE native /api/mix/item_list/ page (<= 20 videos, under `videos`) per call, paged through TikTok's own offset. Any ONE of playlist_ids / usernames / user_ids on the first call (the first given wins, in that order); ignored when `cursor` is supplied. - usernames (optional, string): TikTok handle / @handle / profile URL -> profile mode: every playlist of the creator, playlist by playlist in TikTok's order. The frozen playlist plan (ids + counts) rides in the cursor, so each call fetches ONE native page and continues into the next playlist without re-walking; next_cursor is null after the last playlist. - user_ids (optional, string): Numeric TikTok user id -> profile mode (resolved via /share/user/{id}). - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - enrich (optional, string, one of: auto|true|false, default auto, locked in cursor): auto = skip the per-video pages (a native page holds up to 30 videos, above the 10-video auto limit; each page costs ~90 KB of proxy traffic); true = open every video's page for location_created, diversification_labels, aigc flags; false = never (fastest). Best-effort inside the call's 70 s budget: if TikTok rate-limits video pages or time runs out, rows still return (enriched=false) with a `warning`. - strict_owner (optional, boolean, default false, locked in cursor): true = playlist URL @handle != real owner -> 404. - region (optional, string, default US, locked in cursor): 2-letter market region for the signed API calls. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. No parameter is marked required; send at least one input (see descriptions). Example: curl "https://api.scrapier.io/api/v2/tiktok-playlist-videos/" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-playlist-videos.md ## GET /tiktok-profile/ TikTok Profile Videos: Structured TikTok data: send `usernames`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.videos · Long-running (client timeout 210 s or more) Parameters: - usernames (required, string, example mrbeast): TikTok username (without @). Required on first call; ignored when `cursor` is supplied. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-profile/?usernames=mrbeast" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-profile.md ## GET /tiktok-profile-region/ TikTok Profile Region: Structured TikTok data: send `usernames`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.profile Parameters: - usernames (required, string, example shakira): TikTok handle or profile URL: "shakira", "@shakira" or "https://www.tiktok.com/@shakira". One profile per call (single lookup, next_cursor is always null). - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - videos_to_check (optional, integer, default 5): Newest public videos whose locationCreated vote for the region (0-13, default 5). 0 = author card only, region null. - include_videos (optional, boolean, default true): Attach recent_videos (the per-video evidence behind the region vote). - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-profile-region/?usernames=shakira" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-profile-region.md ## GET /tiktok-profile-video/ TikTok Profile Video: Structured TikTok data: send `usernames`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.videos Parameters: - usernames (required, string, example mrbeast): TikTok handle (mrbeast / @mrbeast), profile URL or numeric user_id. Page 1 = the profile's newest videos (embed window, pinned first, <= 13); later pages walk the creator's playlists with TikTok's own playlist / item_list cursors (~15-20 per page). Rows under `videos`, each with the flat summary + post / computed / aweme blocks; the envelope (request, author, pagination, meta) is at the top level. Each call is bounded (~60 s budget): a slow upstream returns partial rows + next_cursor + `warning`. next_cursor is null once the guest-reachable pool (embed window + playlists) is exhausted. Required first call; ignored when `cursor` is supplied. - user_id (optional, string): Numeric TikTok user id (resolved via /share/user/{id}); wins over a handle, which is cross-checked. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - max_cursor (optional, string, locked in cursor): Optional filter: ms unix time -> only videos created strictly before it are returned (applied to every page). - sort_by (optional, string, one of: latest|popular, default latest, locked in cursor): Compatibility only (no reorder; sort_applied false). - trim (optional, boolean, default false, locked in cursor): Compatibility flag, echoed in request. - complete_schema (optional, boolean, default true, locked in cursor): true = full native mobile-API aweme schema (~184 keys); false = web-exposed fields only. - discovery_limit (optional, integer, default 200000, locked in cursor): Max playlist videos walked over the whole cursor chain; 0 = embed window only (newest 13, one page). - enrich (optional, string, one of: auto|true|false, default auto, locked in cursor): auto = open the video pages (likes / comments / music / hashtags) of the first 3 page-1 (embed) videos only, the rest carry embed data (views, caption, cover, play URL); true = every page-1 video and also every playlist video; false = never (page 1 then carries embed data only: views, caption, cover, play URL). - region (optional, string, default US, locked in cursor): 2-letter country code (priority_region / viewer_region; {region} proxy placeholder only if the platform proxy URL has one). - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-profile-video/?usernames=mrbeast" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-profile-video.md ## GET /tiktok-search-suggestions/ TikTok Search Suggestions: Structured TikTok data: send `queries`. Single lookup, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: single lookup · Results: data.suggestions Parameters: - queries (required, string, example dogs): Partial search term typed into TikTok's search box -> its whole autocomplete list (<= 20 rows, TikTok's rank order) under `suggestions` in ONE guest request. The source has no paging token, so next_cursor is always null. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - region (optional, string, default us, locked in cursor): ISO 3166-1 alpha-2 country to localise the list (default us): carrier / system region + that market's UI language; US -> TikTok's US datacenter host first. Echoed on the envelope; pins the proxy exit country. - language (optional, string, locked in cursor): UI language override layered on top of region, e.g. pt-BR. Blank = the market's default. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-search-suggestions/?queries=dogs" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-search-suggestions.md ## GET /tiktok-search-users/ TikTok Search Users: Structured TikTok data: send `queries`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.users · Long-running (client timeout 210 s or more) Parameters: - queries (required, string, example cooking): Search keyword -> the accounts TikTok's user search returns (guest). One native page (10 accounts) per call. Required first call, ignored when `cursor` is supplied. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - include (optional, string, one of: |profile, locked in cursor): 'profile' hydrates every row of the page (bio, ISO region, bio_link, business_category, following / video counts). - country (optional, string, locked in cursor): ISO 3166-1 alpha-2: only accounts whose public region matches; the page then holds up to max_items (1..20, default 8) in-country accounts, region-checking up to country_scan_limit (10..60, default 20) candidates per call. - max_items (optional, integer, default 8, locked in cursor): Country mode only: in-country page size (1..20, default 8). Ignored otherwise (the search API serves 10 per page). - region (optional, string, default US, locked in cursor): Search market (ISO-2) sent to TikTok. Default: country, else US. - trim (optional, boolean, default false, locked in cursor): true drops the always-null wrappers around each native user_info row. - country_scan_limit (optional, integer, default 20, locked in cursor): Country mode: candidates region-checked per call. - region_samples (optional, integer, default 2, locked in cursor): Newest videos checked per account for the region vote. - hydrate_passes (optional, integer, default 3, locked in cursor): Retry passes for rows left incomplete by transient rate limits. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-search-users/?queries=cooking" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-search-users.md ## GET /tiktok-shop-data-scraper/ TikTok Shop Data Scraper: Structured TikTok data: every input is optional. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.products Parameters: - queries (optional, string, default shoes): TikTok Shop product search keyword (keyword mode). Required first call unless a url is given. Aliases: searchKeywords, query, q. - urls (optional, string): Product /pdp/, category /c/, or seller /store/ URL (URL mode — takes precedence over queries). Aliases: startUrls. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - include_reviews (optional, boolean, default false, locked in cursor): Fetch embedded product reviews. - max_details (optional, integer, default 4): Products per call that get the full product-page enrichment (creators, stock, reviews); the rest are returned from listing data (source="listing"). Each enriched product costs ~65 KB of proxy traffic. 0 = listing only; raise (e.g. 100) to enrich everything. Defaults to all when include_reviews is true. - max_reviews (optional, integer, default 10, locked in cursor): Max reviews per product (1-100). - sort_by_sold_count (optional, string, one of: none|highest_first|lowest_first, default none, locked in cursor): none | highest_first | lowest_first. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. No parameter is marked required; send at least one input (see descriptions). Example: curl "https://api.scrapier.io/api/v2/tiktok-shop-data-scraper/" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-shop-data-scraper.md ## GET /tiktok-shop-product-review/ TikTok Shop Product Review: Structured TikTok data: send `urls`. Single lookup, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: single lookup · Results: data.product Parameters: - urls (required, string, example https://www.tiktok.com/shop/pdp/cat-nail-clipper-by-potaroma-adjustable-sizes-built-in-file-safe-for-kittens-cats/1731578642912612516): TikTok Shop PDP URL. Single lookup -- next_cursor null (browser-signed pagination runs inside one session per call; see module docs). - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - region (optional, string, default US, locked in cursor): - max_items (optional, integer, default 30, locked in cursor): - browser_pagination (optional, boolean, default false, locked in cursor): false (default) = SSR first page only; true opts into heavy browser pagination. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-shop-product-review/?urls=https%3A%2F%2Fwww.tiktok.com%2Fshop%2Fpdp%2Fcat-nail-clipper-by-potaroma-adjustable-sizes-built-in-file-safe-for-kittens-cats%2F1731578642912612516" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-shop-product-review.md ## GET /tiktok-shop-search/ TikTok Shop Search via EchoTik: Structured TikTok data: send `queries` and `country_code`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.products Parameters: - queries (required, string, example cloth): TikTok Shop product search keyword via EchoTik. Required first call. Aliases: keyword, q. - country_code (required, string, one of: US|VN|TH|PH|MY|ID|GB|SG|ES|MX|DE|IT|FR|BR|JP, example US, locked in cursor): Target market. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - order (optional, string, default total_sale_nd_cnt, locked in cursor): EchoTik sort key. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-shop-search/?queries=cloth&country_code=US" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-shop-search.md ## GET /tiktok-single-video-info/ TikTok Single Video Info: Structured TikTok data: send `urls`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.video Parameters: - urls (required, string, example https://www.tiktok.com/@scout2015/video/6718335390845095173): TikTok video / photo URL, vm./vt./tiktok.com/t/ short link, legacy m.tiktok.com/v/.html or bare 19-digit id. Single lookup: the full record is returned under `video`, next_cursor is always null. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - include_transcript (optional, boolean, default true): Fetch + parse the WebVTT caption track (`transcript` text + `transcript_info`). - include_media_urls (optional, boolean, default true): Attach `download_media_urls` (no-watermark mp4 / watermarked mp4 / carousel images as CDN URLs with the `fetch_headers` to download them). - trim (optional, boolean, default false): true = condensed aweme_detail instead of the full 188-key record. - complete_schema (optional, boolean, default true): Pad aweme_detail to the full native mobile-API schema. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-single-video-info/?urls=https%3A%2F%2Fwww.tiktok.com%2F%40scout2015%2Fvideo%2F6718335390845095173" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-single-video-info.md ## GET /tiktok-top-search/ TikTok Top Search: Structured TikTok data: send `queries`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.items Parameters: - queries (required, string, example dance challenge): Search keyword for the TikTok search Top tab (videos + photo carousels). One keyword per call; required first call, ignored when `cursor` is supplied. One native TikTok page (~30 rows) per call; next_cursor null when TikTok has no more (guest Top search is shallow: ~60-100 rows per keyword). - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - publish_time (optional, string, one of: all-time|yesterday|this-week|this-month|last-3-months|last-6-months, default all-time, locked in cursor): Upload-date filter (raw 0/1/7/30/90/180 accepted). - sort_by (optional, string, one of: relevance|most-liked|date-posted, default relevance, locked in cursor): Ranking (raw 0/1/3 accepted). - region (optional, string, locked in cursor): ISO-2 search market sent to TikTok and, through the platform proxy, the exit country (default US). Does NOT filter rows - use filter_region. - include_region (optional, boolean, default false, locked in cursor): true = load the video page of each row (~90 KB proxy traffic each) for post.ext.region (+ CDN mirrors, subtitles). Default false: fast, region null. - filter_region (optional, string, locked in cursor): Client-side ISO-2 filter on post.ext.region (forces include_region). A page can be empty with a non-null next_cursor - keep following it. - count (optional, integer, default 30, locked in cursor): TikTok page size 1..50 (default 30). - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-top-search/?queries=dance+challenge" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-top-search.md ## GET /tiktok-transcript/ TikTok Transcript: Structured TikTok data: send `urls`. Single lookup, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: single lookup · Results: data.transcript · Long-running (client timeout 210 s or more) Parameters: - urls (required, string, example https://www.tiktok.com/@natgeo/video/7683891628230315295): TikTok video / photo URL, vm./vt./tiktok.com/t/ short link, legacy m.tiktok.com/v/.html or bare 19-digit id. Single lookup: the full record (transcript text + VTT + segments, every caption track inline, moments, post metadata) is returned under `transcript`, next_cursor is always null. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - language (optional, string, default en, locked in cursor): 2-letter code of the transcript wanted; falls back to the original-language track. en on a non-English video without captions is translated by Whisper. - use_ai_as_fallback (optional, boolean, default true, locked in cursor): faster-whisper transcription for videos WITHOUT caption tracks. - ai_max_duration_seconds (optional, integer, default 60, locked in cursor): Longest video the AI fallback transcribes on default 60, hard cap 90. Longer videos without captions return transcript null + transcript_reason. - transcribe_all (optional, boolean, default false, locked in cursor): AI-transcribe even when captions exist (captions still returned). - moments (optional, string, one of: true|false|brand|sponsor_read|product_demo|call_to_action, default true, locked in cursor): true/all, false, or a CSV of moment kinds located on the transcript timeline. - brand (optional, string, locked in cursor): Brand / product name for the `brand` moment kind. - whisper_model (optional, string, one of: tiny|base|small, default base, locked in cursor): faster-whisper size for the AI path (base pre-warmed). - whisper_device (optional, string, one of: cpu|auto|cuda, default cpu, locked in cursor): Inference device for faster-whisper. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-transcript/?urls=https%3A%2F%2Fwww.tiktok.com%2F%40natgeo%2Fvideo%2F7683891628230315295" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-transcript.md ## GET /tiktok-trending-hashtag/ TikTok Trending Hashtag: Structured TikTok data: every input is optional. Cursor-paginated, 2 credits per successful call. Cost: 2 credits per successful call · Pagination: cursor · Results: data.data Parameters: - country (optional, string, one of: US|FR|DE|IT|ES|GB|AR|AU|BR|CA|CO|EG|ID|IL|JP|KR|MY|MX|PH|SA|SG|ZA|TW|TH|TR|AE|VN, default US): TikTok ranking country. One country per request; defaults to US. - period (optional, integer, one of: 7|30|90|120, default 30): Ranking window in days. Legacy 120 maps to 90. - result_type (optional, string, one of: top100|top100_with_analytics|hashtag_analytics, default top100): top100 is ranking-only. Analytics modes add trend and optional challenge metrics; anonymous source does not guarantee 100 results. - industry_id (optional, string): Optional top-level TikTok industry ID. Restricts the feed to that industry. - include_industries (optional, boolean, default false): Expand overall ranking with industry rankings. Rank remains relative to each industry; more requests. - related_videos_source (optional, string, one of: none|tikwm, default none): Optional TikWM third-party related videos in analytics modes. Adds requests; no cookies. - related_items_limit (optional, integer, default 3): Related videos per hashtag when TikWM is enabled. Deadline may return fewer with a warning. - related_items_cursor (optional, integer, default 0): Optional TikWM starting cursor for related videos. - cursor (optional, string): Opaque continuation token. Cursor-only calls preserve country, window, industry, and analytics options. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. No parameter is marked required; send at least one input (see descriptions). Example: curl "https://api.scrapier.io/api/v2/tiktok-trending-hashtag/" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-trending-hashtag.md ## GET /tiktok-trending-videos/ TikTok Trending Videos: Structured TikTok data: send `country_code`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.videos Parameters: - country_code (required, string, one of: US|JP|VN|TH|ID, example US): Country: US, JP, VN, TH or ID (the countries TikTok Creative Center ranks videos for). Required on first call; ignored when `cursor` is supplied. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - sort_by (optional, string, one of: hot|views|engagement|six_second_views|view|like|share|comment, default hot): Ranking metric: hot/views (video views), engagement (alias like/share/comment), six_second_views. Locked into the cursor. - period (optional, string, one of: 7|30|weekly|monthly, default 7): 7 (videos created in last 7 days), 30 (last 30 days), weekly, monthly. Deeper pages walk older ranking windows. Locked into the cursor. - content_tag (optional, string): Optional content filter (comma-separated): beauty, fashion, food, lifestyle, tech. Empty = all. Locked into the cursor. - organic_only (optional, boolean, default false): Only organic (non-promoted) videos. Locked into the cursor. - page_size (optional, integer, default 25): Videos per page (1-100). Locked into the cursor. - include_details (optional, boolean, default false): true = add tiktok.com stats (likes, shares, comments, saves, plays, duration) per video, fetched in parallel within the call ~24 s budget; a video whose stats miss the budget is still returned with stats null and stats_pending=true (Creative Center views/engagement always present). false = listing only (~3-5 s). Locked into the cursor. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-trending-videos/?country_code=US" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-trending-videos.md ## GET /tiktok-user-liked/ TikTok User Liked: Structured TikTok data: send `usernames`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.items Parameters: - usernames (required, string, example chipotle): TikTok username (chipotle / @chipotle), profile URL or numeric user id -> the videos this account has LIKED, newest like first: ONE native /api/favorite/item_list/ page (30 liked videos, under `items`) per call, paged through TikTok's own like-time cursor (+ the boundary ids it repeats, so pages never overlap). Required on the first call; ignored when `cursor` is supplied. Hidden / private liked list -> 404 "likes are private"; unknown handle -> 404; TikTok refusing every attempt -> 502. - user_ids (optional, string): Numeric TikTok user id (6646487634225135622), resolved via /share/user/{id}. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - enrich_pages (optional, string, default false, locked in cursor): false (default) = liked-list values only (fastest); true = one video page load per liked video (precise duration, region, TikTok labels, pinned flag; best effort inside the call deadline, ~10-20 s slower); N = only the first N rows of each page. - region (optional, string, locked in cursor): ISO-2 market for the signed API calls (default: the page's region, else US); pins the proxy exit country. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-user-liked/?usernames=chipotle" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-user-liked.md ## GET /tiktok-users-audience-demographics/ TikTok Users Audience Demographics: Structured TikTok data: send `usernames`. Cursor-paginated, 3 credits per successful call. Cost: 3 credits per successful call · Pagination: cursor · Results: data.demographics Parameters: - usernames (required, string, example shakira): TikTok handle or profile URL: "shakira", "@shakira" or "https://www.tiktok.com/@shakira". One profile per call (single lookup, next_cursor is always null). - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - sample_size (optional, integer, default 40): Recent followers to sample (default 40). v2 runs synchronously, so it is CAPPED at 40 here and the whole lookup has a 40s deadline: if it runs out, the followers located so far are returned with status "partial" + a `warning`. Use v1 (async) for up to 500 (richer sample). - sample_spread_hours (optional, integer, default 12): Spread the sample over the last N hours of follows; 0 = most recent followers first. - videos_per_follower (optional, integer, default 1): Latest public videos checked per follower (1-5); >1 = majority vote. Multiplies runtime. - max_countries (optional, integer): Cap the audience_locations rows (sorted by count desc). Empty = all. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktok-users-audience-demographics/?usernames=shakira" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktok-users-audience-demographics.md ## GET /tiktoks-by-song/ TikToks by Song: Structured TikTok data: send `clip_ids`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.videos · Long-running (client timeout 210 s or more) Parameters: - clip_ids (required, string, example 7439295283975702544): TikTok sound / song clip id (7439295283975702544 = SZA "luther") or a music page URL -> ONE native TikTok page per call: the 12 videos of TikTok's "popular" order at the cursor's offset (page 1 also carries the artist-pinned video in front = 13 rows), each with the full native aweme + the sound block. Alternatively ONE video URL / vm./vt. short link -> single-video lookup (one row, next_cursor null). Required on the first call; ignored when `cursor` is supplied. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - enrich_pages (optional, boolean, default false, locked in cursor): true = one video page load (~100 KB) per listed video -> region, precise durations, captions, stickers, diversification labels (default false = faster). - include_artist_video (optional, boolean, default true, locked in cursor): Page 1 only: the newest post of a credited artist that uses the sound in front (is_pinned_by_artist); never repeated on later pages. - include_artist_profile (optional, boolean, default true, locked in cursor): Artist / owner profile pages -> verified flag + follower counts in the sound block. - region (optional, string, locked in cursor): ISO-2 market sent to the music API (also pins the proxy exit country; default US). - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/tiktoks-by-song/?clip_ids=7439295283975702544" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/tiktok/tiktoks-by-song.md