Douyin Video Downloader & Scraper - No Watermark
Pricing
Pay per usage
Download Douyin (抖音) videos without watermark and scrape profiles: captions, likes, comments, views, music & hashtags. Export to JSON, CSV, or Excel. No login, no code.
Pricing
Pay per usage
Rating
0.0
(0)
Developer
Harish Garg
Maintained by CommunityActor stats
0
Bookmarked
3
Total users
3
Monthly active users
an hour ago
Last modified
Categories
Share
Douyin video URLs or share links (e.g. https://www.douyin.com/video/123... or https://v.douyin.com/xxxx/). Each produces one row with the no-watermark MP4 URL.
Douyin user profile URLs (https://www.douyin.com/user/MS4wLjABAAAA...). The actor lists each user's posts, newest first.
Search Douyin by keyword and get one row per matching video. Chinese, English, hashtags (e.g. #美食), brand or person names, and emoji all work. Each keyword is searched independently; duplicates across keywords are removed automatically so a video is only returned (and billed) once.
Upper bound on videos fetched per keyword. The actor pages through results until it reaches this number or the keyword runs out of new videos. One row = one billable result, so set this to what you actually need.
How Douyin orders search results. 'Relevance' is the standard search ranking.
Limit search results to videos published within this window.
Limit search results to videos in this length range.
Douyin gates keyword search behind login, so a logged-in cookie is required to use Search keywords. Video URLs, profile URLs, and comments do not need it and keep working without a cookie.
How to get it: log in at douyin.com in your browser, open DevTools → Network, click any request to douyin.com, and copy the full Cookie request header. Paste it here.
Leave empty if you're only scraping URLs. A cookie set here overrides the actor's DOUYIN_COOKIE environment secret if one is configured.
Upper bound on posts fetched per profile URL.
Profile listings don't return like/comment/share/collect counts. Enable to fetch them per video — one extra request each (slower, more billable events). Search results already include these counts, but enabling this also fills in the detail-only seoOcrContent (Douyin's OCR of on-screen text) for search rows. Has no effect on individual Video URLs, which already return full detail.
Fetch each video's top-level comments (text, author, like count, reply count, IP region) and attach them to the row as a comments array. Adds paginated requests per video — slower and more billable. Applies to both Video URLs and Profile URLs.
Upper bound on top-level comments fetched per video when 'Include comments' is on. Fetched newest-first in pages of 20.
Download each item's media file into the run's key-value store and expose it in the dataset as downloadedFileUrl — a permanent, non-expiring link you can use instead of the signed douyinvod.com CDN URLs, which expire after a while. On by default. Turn off to skip the download and rely only on the direct CDN URLs (videoUrlNoWatermark, videoMirrors), which are always returned regardless.
Which file to save when 'Save media file to storage' is on. 'Video' stores the no-watermark MP4 (.mp4). 'Post audio / music track' stores the sound attached to the post as an MP3 (.mp3) — this is the post's music/original-sound resource, NOT a demux of the video's own audio track, so for videos set to a licensed song you get that song, not the in-video audio.
Only applies when 'Save media file to storage' is on. Douyin's CDN links fail intermittently; the actor already retries across mirror URLs. If a file still can't be stored after retries, this marks the whole run as FAILED instead of finishing green — so a missing file is never reported as success. The dataset row is still written with the direct CDN URLs (videoUrlNoWatermark, videoMirrors) and a downloadError field. Turn off to keep the run green and just flag failures per row.
