Kuaishou Scraper | Videos, Comments and Creator Stats avatar

Kuaishou Scraper | Videos, Comments and Creator Stats

Pricing

from $5.95 / 1,000 videos

Go to Apify Store
Kuaishou Scraper | Videos, Comments and Creator Stats

Kuaishou Scraper | Videos, Comments and Creator Stats

Kuaishou (快手) scraper for public videos by link, ID or keyword: views, likes, comments, saves, hashtags, publish time, creator followers and top comments with the commenter's province. 2 links return 22 videos in 17 seconds, 30 fields each. No login, no cookie. Export to JSON, CSV or Excel.

Pricing

from $5.95 / 1,000 videos

Rating

0.0

(0)

Developer

SilentFlow

SilentFlow

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

2 days ago

Last modified

Share

Kuaishou Scraper (快手) | Videos, Comments and Creator Stats (2026)

2 URLs. 22 Kuaishou videos. 30 fields per video.

From SilentFlow. Creators of Pinterest Scraper (#1 for “pinterest pins scraper” in Apify Store search) and Leboncoin Scraper Ppe, with 230,000+ combined lifetime runs.

Turn the Kuaishou links on your campaign sheet into a table: views, likes, comments and saves for every video, the creator's followers on the same row, and what the top commenters say, province by province. 22 videos in under 35 seconds on the prefilled input, no Kuaishou account, no Chinese phone number.

Why choose this actor?

Copying view counts out of the Kuaishou app one video at a time? Picking creators from follower screenshots nobody can check? Using a scraper that wants a numeric photo_id when all you have is the link your agency sent you?

  • 31 filled fields per video, where the most used Kuaishou scraper shows 15 keys. On the prefilled run (22 videos, 28 Sep 2026), 31 fields carried a value on at least half the rows and 28 on every single row. Beyond views, likes, comments and saves you get the creator's followers, following and number of videos (authorFollowers, authorFollowing, authorVideoCount), the hashtags as a clean list, the duration in seconds, the soundtrack, the cover and the video file, and the top comments with the province each one was posted from (authorRegion). On that run 21 of the 22 rows carried 10 top comments and the other one 6.
  • Paste the link you already have, or an ID, or a topic. Web links (kuaishou.com/short-video/...), short share links (v.kuaishou.com/...), share pages and the whole share text copied from the app all work in the same field, next to video IDs and keywords. The most used Kuaishou scraper takes IDs and keywords but no link. On our test with a share page, two share IDs, a numeric photo ID and a profile link, the 4 videos came back in 11 seconds and the profile link was named in the log as not a video.
  • A form of 7 fields, none of them required. The most used Kuaishou scraper asks for 10. Here two videos of large food creators are prefilled with 10 related videos each: press Start and 22 rows come back. Every field has a sensible default, and the comments, the related videos and the keyword cap are one number each.

Try Kuaishou Scraper

Try your first Kuaishou video table

Paste this input into the Input tab (JSON view) and press Start. It is the prefilled input of the form: two home-cooking videos from creators with over a million followers, each followed by the 10 videos Kuaishou recommends next to it, with the 10 top comments of every video.

{
"videoUrls": [
"https://www.kuaishou.com/short-video/3xwdz5sxnu5fat9",
"https://www.kuaishou.com/short-video/3xqvt79p7gewvbg"
],
"maxItems": 50,
"relatedVideos": 10,
"maxComments": 10,
"debugMode": false
}

Runs of this size fit easily in the Apify Free plan, so you can try the actor on your own links before scheduling anything. Each video is one row of the dataset; export it as JSON, CSV, Excel or HTML from the Storage tab, or read it from the API. The Videos view shows caption, creator, followers, views, likes, comments, publish date and link at a glance; the Creators view lists each creator with their Kuaishou ID, gender, followers, following, number of videos and profile link.

How it works

Copy to your AI assistant

Paste this block into Claude, ChatGPT or Cursor to give it full context about this actor:

You have access to the Kuaishou Scraper on Apify: silentflow/kuaishou-scraper
It returns one row per public Kuaishou (快手) video. No Kuaishou account is needed.
Input (all optional, at least one of videoUrls, videoIds or keywords):
- videoUrls (array of strings): Kuaishou video links. Accepts https://www.kuaishou.com/short-video/<id>,
short share links https://v.kuaishou.com/<code>, share pages https://m.gifshow.com/fw/photo/<id>
and the share text copied from the app. Profile and live links are skipped.
- videoIds (array of strings): share IDs such as "3xwdz5sxnu5fat9" or numeric photo IDs such as "5211509402781314482".
- keywords (array of strings): topics, Chinese words work best ("美食", "游戏", "成都 美食").
Keeps the recommended videos, and the videos recommended next to each match, whose caption
or hashtags contain every word. Broad topics fill fast, rare words may find few videos.
- maxItems (integer, default 50): videos saved per keyword. Links and IDs are always all read.
- relatedVideos (integer, default 0, max 200): for each video read by link or ID, also save this many
of the videos Kuaishou recommends next to it (relatedTo names the source video).
- maxComments (integer, default 10, max 20): top comments saved on each row, 0 to skip.
- debugMode (boolean, default false).
Output per video (34 top-level fields, absent values omitted):
- id (string, share ID), photoId (string, numeric ID), url (string, permanent video page)
- caption (string), hashtags (array of strings), type (string: video or image),
durationSeconds (number), width (integer), height (integer), musicName (string)
- authorId (string), authorUserId (string), authorName (string), authorKwaiId (string, when set),
authorGender (string: female or male), authorAvatarUrl (string), authorUrl (string, profile page),
authorFollowers, authorFollowing, authorVideoCount (integers), authorVerified (boolean)
- viewCount, likeCount, commentCount, collectCount (integers), shareCount (integer, when Kuaishou shows it)
- publishedAt (string, RFC 3339 UTC)
- coverUrl (string), videoUrl (string, mp4 link that expires)
- topComments (array of {id, text, authorName, authorId, authorRegion, likeCount, replyCount, publishedAt})
- source (string: url, id, keyword or related), keyword (string), relatedTo (string), scrapedAt (string, RFC 3339 UTC)
Use apify-client for Python or JavaScript: client.actor("silentflow/kuaishou-scraper").call(run_input=...)

How to scrape Kuaishou videos

Three runs, from the simplest to the richest. Each JSON is a complete input: paste it into the JSON view of the Input tab.

Copy the share link from the Kuaishou app (or the whole share text, the link is found inside it) and paste it. One row comes back with the video, its creator and its 10 top comments.

{
"videoUrls": ["https://www.kuaishou.com/short-video/3xwdz5sxnu5fat9"]
}

Report on a list of campaign videos without the comments

Paste the links or IDs your agency sent you, in any mix. Set maxComments to 0 when you only need the numbers: the run is faster and each row still carries views, likes, comments, saves and the creator's audience.

{
"videoUrls": [
"https://m.gifshow.com/fw/photo/3x55xca9nfrd2n2",
"https://www.kuaishou.com/short-video/3xqvt79p7gewvbg"
],
"videoIds": ["3x7g2d2a5mn5yz9", "5211509402781314482"],
"maxComments": 0
}

Collect 50 food videos and the talk around them

Type a topic in Chinese. The actor keeps the recommended videos whose caption or hashtags mention it, then the videos Kuaishou shows next to each match, until it has 50. On our test, a 20-video run on 美食 returned 20 videos that all had 美食 in their caption, in 44 seconds.

{
"keywords": ["美食"],
"maxItems": 50,
"maxComments": 10
}

Input parameters

FieldTypeDefaultDescription
videoUrlsarray of stringsprefilled with 2 linksKuaishou video links, one per line. See the accepted formats below. Each link becomes one row.
keywordsarray of stringsnoneTopics to find videos about. Chinese words find the most. Every word of a keyword must appear in the caption or hashtags.
videoIdsarray of stringsnoneShare IDs (3xwdz5sxnu5fat9) or numeric photo IDs (5211509402781314482). Each ID becomes one row.
maxItemsinteger50Videos saved for each keyword. Videos read by link or ID are always all saved and do not count.
relatedVideosinteger0 (prefilled 10)For each video read by link or ID, also save this many of the videos Kuaishou recommends next to it. Up to 200.
maxCommentsinteger10Top comments saved in topComments on every row, up to 20. 0 skips them.
debugModebooleanfalseAdds detailed lines to the run log.

No field is required: a run needs at least one link, ID or keyword, and the form comes with two links prefilled.

Accepted formats

  • Web links: https://www.kuaishou.com/short-video/3xwdz5sxnu5fat9, with or without the parameters Kuaishou adds (?authorId=...&streamSource=...), and the older https://www.kuaishou.com/photo/<creator>/<video> form.
  • Short share links: https://v.kuaishou.com/2Ccihtp and https://www.kuaishou.com/f/X-.... The link is opened to find the video it points to.
  • Share pages: https://m.gifshow.com/fw/photo/<id>, https://c.kuaishou.com/fw/photo/<id>, https://v.m.chenzhongtech.com/fw/photo/<id> and long-video pages /fw/long-video/<id>.
  • Share text from the app: paste the whole text ("看看这个作品 https://v.kuaishou.com/... 复制此链接"), the link is found inside it.
  • IDs: the share ID at the end of a link, or the numeric photo ID. Upper or lower case both work.
  • Not videos: profile links (/profile/..., /fw/user/...) and live rooms are skipped and named in the log. A short link that turns out to open a profile is skipped the same way.
  • The same video given twice (as a link and as an ID, say) is saved once.

What data can you extract from Kuaishou?

Every row is one video, with up to 34 fields in seven groups. Numbers are numbers, the publish time is RFC 3339 in UTC, IDs are strings (Kuaishou's numeric IDs are too long for some spreadsheets to keep as numbers), and a value Kuaishou does not show is left out of the row rather than written as an empty string or a made-up zero.

GroupFieldsWhat they tell you
Identity (3)id, photoId, urlThe share ID used in links, Kuaishou's numeric ID, and the permanent web page of the video.
Content (7)caption, hashtags, type, durationSeconds, width, height, musicNameThe text as posted, its hashtags as a list, video or image post, length in seconds, frame size (portrait or landscape) and the soundtrack name.
Creator (11)authorId, authorUserId, authorName, authorKwaiId, authorGender, authorAvatarUrl, authorUrl, authorFollowers, authorFollowing, authorVideoCount, authorVerifiedWho posted it and how big the account is: followers, following, number of videos, the Kuaishou ID when the creator set one, and the profile link.
Measures (5)viewCount, likeCount, commentCount, collectCount, shareCountViews, likes, comments and saves on every video; shares when Kuaishou displays them (on 6 to 8 of the 22 videos of our runs).
Time (1)publishedAtWhen the video went live, in UTC.
Media and comments (3)coverUrl, videoUrl, topCommentsThe cover image, the mp4 file, and the top comments: text, author, the province it was posted from, likes, replies and date.
Meta (4)source, keyword, relatedTo, scrapedAtHow the row was found (link, ID, keyword or related video), the keyword or the video it came from, and when it was read.

url, authorUrl, id and photoId are permanent. coverUrl and videoUrl are CDN links: download what you need to keep soon after the run, since media links on Kuaishou expire. The comment region (authorRegion) is the province Kuaishou shows under each comment, such as 广西, 江西 or 黑龙江.

Output example

One complete row from the prefilled run (28 Sep 2026), with topComments cut to 2 of its 10 comments for reading:

{
"id": "3xqvt79p7gewvbg",
"photoId": "5219672176234957753",
"url": "https://www.kuaishou.com/short-video/3xqvt79p7gewvbg",
"caption": "哪种才是鸡蛋最好吃的做法? #鸡蛋的神仙吃法 #好好吃饭 #家常菜 #美食教程",
"hashtags": ["鸡蛋的神仙吃法", "好好吃饭", "家常菜", "美食教程"],
"type": "video",
"durationSeconds": 159.3,
"width": 720,
"height": 1280,
"musicName": "李哥我饿了的作品原声",
"authorId": "3xcvc4qrnrfmqiu",
"authorUserId": "2956788831",
"authorName": "李哥我饿了",
"authorAvatarUrl": "https://p22.a.yximgs.com/uhead/AB/2022/08/11/10/BMjAyMjA4MTExMDAxNDFfMjk1Njc4ODgzMV8xX2hkMjY1Xzkx_s.jpg",
"authorUrl": "https://www.kuaishou.com/profile/3xcvc4qrnrfmqiu",
"authorFollowers": 1049006,
"authorFollowing": 5,
"authorVideoCount": 326,
"authorVerified": false,
"viewCount": 454571,
"likeCount": 23048,
"commentCount": 135,
"shareCount": 1615,
"collectCount": 8910,
"publishedAt": "2026-08-30T03:00:00Z",
"coverUrl": "https://p23.a.yximgs.com/upic/2026/08/29/21/BMjAyNjA4MjkyMTQ1MjhfMjk1Njc4ODgzMV8yMDczMjIzMjY0NjZfMF8z_B75c00632e3bffc84b101513a45e0864b.jpg?clientCacheKey=3xqvt79p7gewvbg.jpg&di=5a3cae87&bp=10000",
"videoUrl": "https://hwmov.a.yximgs.com/upic/2026/08/29/21/BMjAyNjA4MjkyMTQ1MjhfMjk1Njc4ODgzMV8yMDczMjIzMjY0NjZfMF8z_b_Bd3fa31e9a9479529c3aacce87f4de6c2.mp4?clientCacheKey=3xqvt79p7gewvbg_b.mp4&tt=b&di=5a3cae87&bp=10000",
"topComments": [
{
"id": "1173551400769",
"text": "水蒸蛋吃起来面面的是为什么?",
"authorName": "星月落",
"authorId": "4572652254",
"authorRegion": "黑龙江",
"likeCount": 63,
"replyCount": 4,
"publishedAt": "2026-09-02T14:26:33Z"
},
{
"id": "1172615093070",
"text": "嫂子咋没出镜[奸笑]",
"authorName": "甜圆圆有福仔",
"authorId": "853173188",
"authorRegion": "陕西",
"likeCount": 31,
"replyCount": 5,
"publishedAt": "2026-08-30T04:21:44Z"
}
],
"source": "url",
"scrapedAt": "2026-09-28T03:23:40Z"
}

This creator did not set a Kuaishou ID, so authorKwaiId is absent from the row; keyword and relatedTo only appear on rows found by a keyword or as a related video.

Advanced usage

Grow one viral video into a niche map

Paste one video that performs in your niche and ask for 50 related videos. Kuaishou recommends videos close to the one you start from: on our test, one link with 20 related videos returned 21 rows in 19 seconds, and the related videos of a home-cooking video came back mostly as recipe and home-cooking videos. Sort by viewCount to see what the niche rewards.

{
"videoUrls": ["https://www.kuaishou.com/short-video/3x55xca9nfrd2n2"],
"relatedVideos": 50,
"maxComments": 5
}

Watch several topics at once

Put one topic per line. Each keyword gets its own maxItems videos, and a video that matches two keywords is saved once, under the first. On our test 游戏 (games) returned 10 videos in 14 seconds.

{
"keywords": ["游戏", "农村", "猫"],
"maxItems": 30,
"maxComments": 5
}

Vet a creator shortlist before a deal

Paste one or two recent videos of each creator you are considering. Every row carries the creator's followers, following and number of videos next to the video's views and likes, so you can compare likes per view and audience size side by side in the Creators view, and read the top comments for red flags.

{
"videoUrls": [
"https://www.kuaishou.com/short-video/3xwdz5sxnu5fat9",
"https://www.kuaishou.com/short-video/3xqvt79p7gewvbg",
"https://www.kuaishou.com/short-video/3x55xca9nfrd2n2"
],
"maxComments": 20
}

Track campaign videos every week on a schedule

Save the input below as a task and schedule it weekly in the Schedules tab. Each run reads the same videos again, so the dataset of each run is a snapshot of views, likes, comments and saves you can chart over time (join the runs on id).

{
"videoIds": ["3xwdz5sxnu5fat9", "3xqvt79p7gewvbg", "3x7g2d2a5mn5yz9"],
"maxComments": 0
}

Rank the videos of a topic by engagement in Python

Pull the rows with the Apify Python client and rank them by likes per view, with the creator's audience next to each video.

from apify_client import ApifyClient
client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("silentflow/kuaishou-scraper").call(
run_input={"keywords": ["美食"], "maxItems": 50, "maxComments": 0}
)
rows = list(client.dataset(run["defaultDatasetId"]).iterate_items())
for r in sorted(rows, key=lambda r: r.get("likeCount", 0) / max(r.get("viewCount", 1), 1), reverse=True)[:10]:
print(r["authorName"], r.get("authorFollowers"), r.get("viewCount"), r.get("likeCount"), r["url"])

FAQ

What is Kuaishou? Kuaishou (快手, "quick hand") is one of China's two largest short-video apps, next to Douyin. It is known for everyday, rural and small-town content, live commerce and very loyal creator communities. Its international version is called Kwai. This actor reads the Chinese Kuaishou.

What does this actor return? One row per public video, with up to 34 fields: the video (ID, link, caption, hashtags, type, duration, size, soundtrack), its numbers (views, likes, comments, saves, and shares when shown), when it was published, the cover and the video file, the creator (name, IDs, gender, avatar, profile link, followers, following, number of videos, verification) and the top comments with the province each was posted from.

Do I need a Kuaishou account or a Chinese phone number? No. The actor reads what anyone sees when they open a shared Kuaishou video. There is no login, no API key and nothing to install.

Which inputs can I give it? Video links in any of the shapes Kuaishou shares (web link, short link, share page, share text from the app), video IDs, and keywords, all in the same run. Rows found by link or ID come first, in the order you pasted them, then the related videos, then the keyword results.

How does the keyword mode find videos? It reads the videos Kuaishou recommends to a visitor, keeps those whose caption or hashtags contain every word of your keyword, and then reads the videos Kuaishou recommends next to each match, which tend to be on the same topic. It is not the ranking of Kuaishou's in-app search box. Broad topics such as 美食 (food), 游戏 (games), 农村 (rural life) or 猫 (cats) fill quickly; a rare or very specific word may find few videos or none, and the run says so. For an exact video, paste its link.

Can I scrape every video of a creator's profile? Not with this actor: Kuaishou shows a creator's full list of videos only to logged-in users, and this actor never asks for an account. Paste the creator's videos you know, each row carries the creator's followers, following and total number of videos.

Why is shareCount missing on some rows? Kuaishou shows the share count on some videos only (6 to 8 of the 22 videos of our runs). When it is not shown, the field is left out of the row instead of being set to zero, so an average of shares is never pulled down by fake zeros.

How fresh is the data? Every run reads Kuaishou live when it runs. Views, likes and comments are the values Kuaishou shows at that moment, and the creator's followers are read again for every video.

Are comments always included? Up to 20 top comments per video, 10 by default. Kuaishou limits how many comments it shows to one reader in a row; when it stops showing them, the next videos are saved with their numbers and without topComments, and the log says so. Set maxComments to 0 when you do not need them.

How long do the media links last? url, authorUrl and the IDs are permanent. coverUrl and videoUrl are temporary CDN links: download the files you want to keep shortly after the run.

Can I run it on a schedule or from my code? Yes. Save an input as a task and schedule it in the Schedules tab, or call it from the API, the Python or JavaScript clients, Make, Zapier or n8n. Each run is a new dataset you can export as JSON, CSV or Excel.

More SilentFlow scrapers

🛒 E-commerce

💼 Jobs

🏠 Real estate

📣 Social media

🏢 Business data

✈️ Travel

🎯 Lead generation

🔎 SEO

Support

Need something this scraper doesn't do yet? We ship features fast.

  • Feature requests go straight to our backlog
  • Enterprise needs? We do custom integrations and high-volume plans
  • Pricing questions? Check the Monetization tab on the actor page

Response time: usually under 24 hours.

This actor extracts publicly available data from Kuaishou (kuaishou.com). It does not bypass any login, paywall or CAPTCHA, and it only reads what any visitor of a shared Kuaishou video can see. Users are responsible for complying with Kuaishou's terms of service and applicable data protection laws (GDPR, CCPA, and PIPL for data about people in China). Rows contain personal data about creators and commenters (names, IDs, avatars, the province shown under a comment): collect only what your purpose needs, store it securely and handle it accordingly. The data returned is informational; check the numbers on Kuaishou before using them in a contract.

Built by SilentFlow for marketers, researchers and data teams who would rather measure Kuaishou than copy it by hand.