Zhihu Scraper - Search 知乎 Questions, Answers, Comments & Users

Scrape Zhihu (知乎) and export answers, articles, questions, comments and user profiles to JSON, CSV or Excel. Search answers and articles by keyword, pull a question's details, collect a question's answers, get an answer's comments, look up a user's profile, or list every answer a user has written - all from one Actor. Each result is one clean, flat row with upvotes, comments, favorites, the full plain-text body and author data.

No login, no cookies, no proxies and no China IP to configure - data is fetched through a fast, managed pipeline. Just pick an operation, add your input and press Start.

Unofficial notice: This is an independent tool and is not affiliated with, endorsed by, or connected to Zhihu, 知乎 or Zhihu Inc. "Zhihu" and "知乎" are trademarks of their respective owners, used here only to describe what the Actor scrapes.

What is Zhihu?

Zhihu (知乎) is China's largest question-and-answer community, often described as the Chinese Quora, with over 100 million monthly users. Its long-form answers and Zhuanlan (专栏) articles cover technology, science, careers, finance, products and everyday life, and are a key source of expert opinion and consumer sentiment in China. This scraper gives you programmatic access to Zhihu's search, questions, answers, comments and users so you can research topics, track opinion and find experts at scale without a Zhihu account.

Operations

Pick one Operation and fill in the matching field:

Operation	Input field	What you get
Search answers & articles by keyword	keywords	Answers and articles matching each keyword, with upvotes, comments, favorites and author data.
Question details	questionUrls	The full question: title, description, answer count, views, followers, comments.
Question answers	questionUrls	The answers to each question (up to the first 200), with full text and engagement.
Answer comments	answerUrls	Top-level comments on each answer, with likes, replies and IP region.
User profile	userUrls	A user's profile: followers, answer and article counts, headline, IP region.
User's answers	userUrls	Every answer a user has written, newest first, paginated.

questionUrls accept zhihu.com/question/... URLs or bare numeric question IDs. answerUrls accept answer URLs (zhihu.com/question/.../answer/...) or bare numeric answer IDs. userUrls accept zhihu.com/people/... URLs or bare user tokens (the part after /people/) - one per line.

How to use it (step by step)

1. Choose an Operation.
2. Fill in the matching input:
   • Search -> add one or more Search keywords (Chinese and English both work).
   • Question details / Question answers -> paste Question URLs / IDs.
   • Answer comments -> paste Answer URLs / IDs.
   • User profile / User's answers -> paste User URLs / tokens.
3. Set Maximum results per input (default 100, or 0 for all available) for the list operations.
4. For question answers, optionally choose Default ranking or Recently updated first. For comments, optionally choose Top or Newest.
5. Press Start. Export the dataset as JSON, CSV, Excel, XML or via the API.

Input

Field	Type	Description
operation	string	Required. search, question_detail, question_answers, answer_comments, user_profile, or user_answers.
keywords	array	Search keywords (for search).
questionUrls	array	Question URLs or IDs (for question_detail, question_answers).
answerUrls	array	Answer URLs or IDs (for answer_comments).
userUrls	array	User URLs or tokens (for user_profile, user_answers).
maxItems	integer	Max results per keyword / question / answer / user. 0 = all available. Default 100.
answerSort	string	default or updated (question answers only).
commentSort	string	score or ts (comments only).

Output

Every row carries a record_type of answer, article, question, comment or user. Common answer and article fields:

Field	Description
content_id, url	Answer or article id and canonical zhihu.com URL.
title, excerpt, content	Question or article title, a short excerpt and the full plain-text body.
voteup_count, comment_count, favorite_count, thanks_count	Engagement metrics.
question_id, question_title, question_url	The question an answer belongs to.
author_id, author_url_token, author_name, author_headline, author_follower_count, author_url	Author data.
created_time, updated_time	Created and last-edited time (ISO-8601 UTC).

Question (question) rows add detail, answer_count, visit_count, follower_count and comment_count. Comment (comment) rows add comment_id, answer_id, like_count, dislike_count, child_comment_count, is_hot and ip_location. User (user) rows add user_id, url_token, name, headline, gender, ip_location, follower_count, answer_count, articles_count and profile_url.

Example answer row:

{
  "record_type": "answer",
  "operation": "question_answers",
  "input": "https://www.zhihu.com/question/2065714833606104237",
  "content_id": "2073124822138341002",
  "url": "https://www.zhihu.com/question/2065714833606104237/answer/2073124822138341002",
  "title": "菲尔兹奖得主谈「人工智能可能会杀死数学」，您如何看待这个问题？",
  "voteup_count": 191,
  "comment_count": 50,
  "favorite_count": 109,
  "thanks_count": 5,
  "question_id": "2065714833606104237",
  "author_name": "Yuhang Liu",
  "author_url_token": "yuhang-liu-34",
  "author_follower_count": 386336,
  "created_time": "2026-08-18T11:11:14+00:00"
}

Common use cases

• Topic and opinion research - see what Zhihu's experts and users say about any product, brand, technology or trend.
• Expert (KOL) discovery - find the top answerers for a niche, then pull their profile and full answer history.
• Brand and competitor monitoring - track questions about a brand, the answers they attract and the comment sentiment.
• Market and AI research - build datasets of long-form Chinese Q&A content, comments and engagement.

Notes on reliability

• Question answers return up to the first 200 answers of a question in Zhihu's default or recently-updated order.
• Search results are Zhihu's own ranking; consecutive result pages can overlap and duplicates are removed automatically, so a keyword may return slightly fewer results than requested.
• Avatar and image URLs are served by Zhihu's CDN and can change; fetch them promptly.
• Upvote, comment and follower counts reflect what Zhihu returns at scrape time.
• If a question, answer or user is deleted, private or restricted, that input is skipped and the run continues.

FAQ

Do I need a Zhihu account, cookies or a proxy? No. Just add your input and press Start.

Can I paste full URLs instead of IDs? Yes - every input field accepts full zhihu.com URLs or bare IDs / tokens. An answer URL pasted into Question URLs uses its question.

Can I run several keywords or URLs at once? Yes - add multiple lines. Cross-input duplicate answers and articles are removed automatically.

Can I try it for free? Yes. Users on the free Apify plan can fetch up to 10 results in total to try the Actor. Upgrade to a paid Apify plan to run it without that limit.

What export formats are supported? JSON, CSV, Excel, XML and HTML, plus the Apify API and integrations such as Google Sheets, Zapier, Make and webhooks.

Pricing

This Actor is pay per result: you are charged for each record it returns (search results, questions, answers, comments and user profiles), plus standard Apify platform usage. Different record types have different prices; see the Pricing tab for current rates. You are never charged for inputs that return nothing.

Related scrapers

• Xiaohongshu Scraper - 小红书 / RedNote notes, users and comments.
• Bilibili Scraper - 哔哩哔哩 videos, creators and comments.
• Reddit Posts Search Scraper - Reddit posts by keyword.