🧠 Zhihu Scraper (知乎): Answers, Questions, Users, Search & Hot

Zhihu scraper (知乎), no login: Zhihu answers to any question, Zhihu questions and answer details, Zhihu articles, Zhihu comments with replies, Zhihu users and followers, Zhihu search and the Zhihu hot list (知乎热榜). No cookies. $2.50 per 1,000 answers.

📋 Table of contents

• What does it do?
• Features
• Pricing
• Input
• Output
• Troubleshooting
• Who uses this Zhihu Scraper?
• Use cases
• FAQ
• API, Python, JavaScript and MCP

What does it do?

Zhihu Scraper collects public data from Zhihu, China's biggest Q&A community. Give it question, answer, article or profile links, or Chinese keywords, and it returns the top answers to questions and every answer by a user, with full text, upvotes, comments and saves, question details (views, followers, answer count), column articles, comment threads with replies, user profiles (followers, answers, upvotes), follower and following lists, content and user search results and the hot list (知乎热榜) with heat.

Features

🎬 What you can scrape on Zhihu

Mode	What you give it	What you get
question_answers	question link or ID	Question answers — top 20 answers to a question
question_detail	question link or ID	Question details — views, followers, answer count
video_detail	video or post link	Answer details — one answer per link
article_detail	article link or ID	Article details — one column article per link
video_comments	video or post link	Answer comments (+ replies)
user_profile	profile link / user ID	User profile — followers, answers, likes
user_posts	profile link / user ID	User answers — all answers by a user
user_articles	profile link / user ID	User articles
followers	profile link / user ID	Followers of a user
following	profile link / user ID	Users a user follows
search_videos	keywords	Search answers, articles & questions (keywords)
search_users	keywords	Search users (keywords)
hot_search	nothing	Hot list (知乎热榜)

✨ Smart features

• Full answer and article text, cleaned from HTML
• The top 20 answers of any question, and every answer a user wrote
• Hot list with heat scores (863万热度 → 8630000)
• Replies are placed right under their comment
• Search keeps real answers, articles and questions and drops ads
• Duplicates are removed automatically
• 🔓 No login, no cookies, no proxy setup
• 💸 Pay only for results: hint and error rows are free

🧭 More scrapers from us

• WeChat Scraper (微信)
• Weibo Scraper (微博)
• RedNote Scraper (小红书)
• Bilibili Scraper
• Douyin Scraper (抖音)

Pricing

You pay per result. No compute or proxy charges on top, and hint and error rows are free.

What you get	Price per 1,000
🧠 Answers and articles from questions, users or search	$2.50
💬 Comments and replies	$0.50
👥 Users from followers, following or search	$1.00
🔍 Single lookups: question, answer, article or profile by link	$1.50
🔥 Hot-list questions	$1.00

How much does it cost to scrape Zhihu data?

Job	Cost
1,000 Zhihu answers	$2.50
10,000 Zhihu answers	$25
10,000 comments	$5.00
Today's hot list (50)	$0.05

Set Max cost per run in the run options and the actor stops cleanly at that amount.

🆓 Free plan

On the free Apify plan each run returns up to 10 results, so you can check the data before you buy. Comment replies need a paid plan. Any paid Apify plan removes the limit.

⚡ Speed

Question and user feeds return about 20 answers per request, so 1,000 answers usually take a minute or two.

Input

Field	What it does
mode	Links, @usernames or IDs: question_answers, question_detail, video_detail, article_detail, video_comments, user_profile, user_posts, user_articles, followers, following. Keywords: search_videos, search_users. No input: hot_search.
urls	One per line: question links (zhihu.com/question/…), answer links, article links (zhuanlan.zhihu.com/p/…), profile links (zhihu.com/people/…) or IDs.
queries	Keywords for the search modes, one per line. Chinese keywords give the best results.
maxItems	Stop after this many rows per link or keyword (comments include their replies).
includeReplies	In video_comments mode, also fetch the replies under each comment that has 3 or more replies (TikTok, Douyin, Bilibili). Replies are billed as comments. Paid Apify plans only.

🧪 Example inputs

Top answers to several questions

{"mode": "question_answers", "urls": ["https://www.zhihu.com/question/2092283931764552084", "https://www.zhihu.com/question/19550225"], "maxItems": 20}

Search answers and articles

{"mode": "search_videos", "queries": ["人工智能", "考研"], "maxItems": 200}

Comments with replies on an answer

{"mode": "video_comments", "urls": ["https://www.zhihu.com/question/2092283931764552084/answer/2092334717102302076"], "maxItems": 300, "includeReplies": true}

Everything a user answered

{"mode": "user_posts", "urls": ["https://www.zhihu.com/people/zhang-jia-wei"], "maxItems": 300}

Today's hot list

{"mode": "hot_search"}

Output

Every row uses the same fields. Export as JSON, CSV, Excel, XML or HTML.

📦 Example output

{
  "platform": "zhihu",
  "item_type": "answer",
  "item_id": "2092334717102302076",
  "url": "https://www.zhihu.com/question/2092283931764552084/answer/2092334717102302076",
  "title": "日媒曝黑客黑掉软银旗下云平台后留下225封勒索信，结果运维找7小时没发现勒索信只是一味重启，如何看待？",
  "description": "牛比啦，继韩国银行泄漏客户信息后，软银又上演魔幻大剧2.0。\n黑客黑进他们旗下的云平台IDC Frontier后，让部分云服务器停机、无法启动，然后把勒索信息放在 /vmfs/vo…",
  "author_id": "yejiming",
  "author_name": "Kitt在进化",
  "author_username": "yejiming",
  "created_at": "2026-10-10T11:24:30+00:00",
  "like_count": 704,
  "comment_count": 46,
  "collect_count": 116,
  "mode": "question_answers",
  "source_input": "2092283931764552084",
  "scraped_at": "2026-10-10T15:41:39.639655+00:00",
  "raw": {
    "…": "full original record"
  }
}

📤 Main fields

Field	Description
platform, item_type, mode, source_input	Where the row came from and what it is (video, user, comment, hint, error…)
item_id, url	Native ID and a link back to the item
title, description, hashtags, created_at	Text and publish time (ISO 8601)
author_id, author_name, author_username	Creator or account
play_count, like_count, comment_count, share_count, collect_count	Engagement counts
raw	The full original record, every field the platform returned
error_message	Explanation on free hint / error rows

Troubleshooting

Following or followers come back empty?

Some users hide these lists. You get a free hint row and no charge.

Getting no results?

Check that the link or username is public and typed correctly. Private accounts, deleted posts and hidden lists return a free hint row that says why. Keyword modes need the Keywords field, link modes need Links.

Getting fewer results than maxItems?

The platform ran out of items for that input (a small profile, a niche keyword), or the run reached its Max cost per run limit. The log says which one.

Some fields are empty?

Platforms don't return every field for every item. Everything the platform did return is kept in raw.

Video or image links stopped working?

Media URLs from these platforms expire after a few hours. Download files right after the run.

The run failed or shows an error row?

Retry once: the platforms sometimes refuse a request for a few minutes. If it keeps failing, open an issue with the run link and we'll fix it fast.

Who uses this Zhihu Scraper?

📊 Market researchers

Read long-form Chinese opinions about brands, products and industries.

🌏 Brands entering China

See what Zhihu users ask and answer about your category.

🎓 Researchers and students

Collect Q&A data and public opinion on any topic.

🧑‍🤝‍🧑 KOL agencies

Find expert creators and check their followers and upvotes.

🤖 AI and data teams

Collect high-quality Chinese long-form answers as training data.

Use cases

Topic research

Pull every answer to the top questions in your category and summarise the arguments.

Trend tracking

Pull the hot list every hour and chart which questions rise.

Expert discovery

Search users by keyword and rank them by followers and upvotes.

FAQ

Can I scrape Zhihu without an account?

Yes. No Zhihu login, Chinese phone number, cookies or API keys.

How many answers do I get per question?

The top 20 answers as Zhihu ranks them. To go wider, add more questions, or use the user answers mode, which returns every answer a user wrote.

Do I get the full answer text?

Yes, cleaned from HTML into plain text.

How much does it cost to scrape 10,000 Zhihu answers?

About $25, at $2.50 per 1,000 answers.

Can I scrape Zhihu comments and replies?

Yes, with the comments mode and Include comment replies on (paid plans).

Can I get the Zhihu hot list (知乎热榜)?

Yes, with heat scores, in one call.

Can I scrape Zhihu articles (知乎专栏)?

Yes: one article by link, or all articles of a user.

Do I need a Chinese proxy?

No.

Is there a free tier?

Yes. On the free Apify plan each run returns up to 10 results. Comment replies need a paid plan.

Can I use it from Python, JavaScript or an AI agent?

Yes, through the Apify API, the client libraries, or MCP.

API, Python, JavaScript and MCP

Python

from apify_client import ApifyClient

client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("openclawai/zhihu-scraper").call(run_input={"mode": "question_answers", "urls": ["https://www.zhihu.com/question/2092283931764552084", "https://www.zhihu.com/question/19550225"], "maxItems": 20})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item)

JavaScript

import { ApifyClient } from "apify-client";

const client = new ApifyClient({ token: "YOUR_APIFY_TOKEN" });
const run = await client.actor("openclawai/zhihu-scraper").call({"mode": "question_answers", "urls": ["https://www.zhihu.com/question/2092283931764552084", "https://www.zhihu.com/question/19550225"], "maxItems": 20});
const { items } = await client.dataset(run.defaultDatasetId).listItems();

MCP (Claude, ChatGPT, Cursor and other MCP clients): https://mcp.apify.com/?tools=openclawai/zhihu-scraper

Works with Zapier, Make, n8n, Google Sheets, Slack, webhooks and scheduled runs.

🛡️ Legal

Collects public data only. You are responsible for complying with Zhihu's terms and with privacy laws such as GDPR, CCPA and PIPL when you store personal data.

⭐ Support

Something not working? Check the hint / error rows, then open an issue with the run link. We reply fast.