# Zhihu Scraper (知乎) | Search, Answers, Comments & Profiles (`zen-studio/zhihu-scraper`) Actor Search Zhihu and collect answers, articles, comments, replies, creator profiles and trending questions. Export structured JSON, CSV or Excel. - **URL**: https://apify.com/zen-studio/zhihu-scraper.md - **Developed by:** [Zen Studio](https://apify.com/zen-studio) (community) - **Categories:** Social media, Automation - **Stats:** 3 total users, 2 monthly users, 100.0% runs succeeded, 0 bookmarks - **User rating**: No ratings yet ## Pricing from $2.99 / 1,000 saved results This Actor is paid per event. You are not charged for the Apify platform usage, but only a fixed price for specific events. Learn more: https://docs.apify.com/actors/running/actors-in-store.md#pay-per-event ## What's an Apify Actor? Actors are web data automations that power AI and operations. They run on the Apify platform to scrape websites, process data, connect APIs, and automate workflows. In Batch mode, an Actor accepts a well-defined JSON input, performs an action which can take anything from a few seconds to a few hours, and optionally produces a well-defined JSON output, datasets with results, or files in key-value store. In Standby mode, an Actor provides a web server which can be used as a website, API, or an MCP server. Actors are written with capital "A". ## How to integrate an Actor? If asked about integration, you help developers integrate Actors into their projects. You adapt to their stack and deliver integrations that are safe, well-documented, and production-ready. The best way to integrate Actors is as follows. - **AI agents and MCP clients** — the [Apify MCP server](https://docs.apify.com/integrations/mcp.md) at `https://mcp.apify.com` (remote, streamable HTTP, OAuth on first use). - **Agentic workflows and local Actor development** — [Agent Skills](https://apify.com/.well-known/agent-skills/index.json) with the [Apify CLI](https://docs.apify.com/cli/docs.md): `npm install -g apify-cli`, then `apify login`. - **JavaScript/TypeScript projects** — the official [JS/TS client](https://docs.apify.com/api/client/js/docs.md): `npm install apify-client`. - **Python projects** — the official [Python client](https://docs.apify.com/api/client/python/docs.md): `pip install apify-client`. - **Any other language** — the [REST API](https://docs.apify.com/api/v2.md). For usage examples, see the [API](#api) section below. For more details, see Apify documentation as [Markdown index](https://docs.apify.com/llms.txt) and [Markdown full-text](https://docs.apify.com/llms-full.txt). # README ## Zhihu Scraper (知乎) | Search, Answers, Comments & Profiles (2026)

Zhihu (知乎), unlocked. Rich data at speed. Built for serious research.

Zhihu (知乎) scraper: questions, answers, authors and engagement as structured JSON

Zen Studio Chinese Social · Research across platforms
知 Zhihu ↗ ➤ You are here	Weibo Search ↗ Posts, authors, engagement	RedNote Search ↗ Notes, images, engagement	Douyin Search ↗ Videos, creators, music

#### Copy to your AI assistant ``` zen-studio/zhihu-scraper on Apify. Collects Zhihu search, content, conversations and creators. Call ApifyClient("TOKEN").actor("zen-studio/zhihu-scraper").call(run_input={...}), then client.dataset(run["defaultDatasetId"]).list_items().items. Choose one operation per run. Full spec: GET https://api.apify.com/v2/acts/zen-studio~zhihu-scraper/builds/default (Bearer TOKEN) → inputSchema, actorDefinition.storages.dataset, readme. Token: https://console.apify.com/account/integrations ``` ### What can this Zhihu scraper collect? · 采集范围 - **Zhihu keyword search · 关键词搜索:** discover answers and articles with content-type, recency, and sorting filters. - **Zhihu answers and comments · 回答与评论:** collect available text, author details, engagement, and optional reply threads from content URLs. - **Zhihu creators and trends · 用户与热榜:** inspect public profiles, published articles, and ranked hot-list questions. | Operation | Input | Results | |---|---|---| | `search` · 内容搜索 | Keywords | Matching answers, articles, profiles, questions, and videos where available | | `question_answers` · 问题回答 | Question URLs or IDs | Available answers with content, authors, and engagement | | `content_detail` · 内容详情 | Answer, article, question, or pin URLs | Individual content records | | `comments` · 评论 | Answer, article, or pin URLs | Top-level comments, optionally including replies | | `comment_replies` · 评论回复 | `comment_id` values from a Comments run | Replies to those comments | | `user_profile` · 用户资料 | Profile URLs or URL tokens | Creator profiles and available activity counts | | `user_content` · 用户文章 | Profile URLs or URL tokens | Published articles only | | `hot_list` · 知乎热榜 | No target required | Ranked trending questions | **Scope:** Question-answer continuation is currently limited. The Actor can collect an initial batch of up to 100 entries, but cannot guarantee a complete answer list. If continuation fails, the run summary marks that target as partial. Creator answer feeds and standalone `/zvideo/` detail URLs are not supported. Pin video URLs are supported. ### How to scrape Zhihu data · 快速开始 1. Open [Zhihu Scraper](https://console.apify.com/actors/YOq90GHMVnTe8qtHO/input) and choose an operation in the Input tab. 2. Enter keywords, supported URLs, or IDs. Set both result limits in the first section; start with 20 total results. 3. Run the Actor, inspect its dataset view, and download JSON, CSV, Excel, or HTML. The same form works without writing code. Developers can submit identical input through the Apify API. #### Search · 内容搜索: collect relevant discussions ```json { "operation": "search", "keywords": ["人工智能"], "max_items": 20 } ``` #### Details · 内容详情: collect an article ```json { "operation": "content_detail", "urls": ["https://zhuanlan.zhihu.com/p/2032860336215307118"], "include_markdown": true } ``` #### Comments · 评论: collect a discussion with replies ```json { "operation": "comments", "urls": ["https://zhuanlan.zhihu.com/p/2032860336215307118"], "comment_sort": "popular", "include_replies": true, "max_items": 20 } ``` Enable **Include replies · 包含回复** to collect comments and their replies in one run. No comment IDs are needed. Leave it off for top-level comments only. Comments and replies share the same total, per-URL, and spending limits. A limit of 20 means at most 20 rows combined, not 20 comments plus extra replies. Each saved row has the same result price. Threads follow their parent comments, so small limits may stop partway through a thread. #### Specific replies · 指定评论回复: use IDs from a Comments run Run **Comments** with a content URL first. In its dataset, select top-level comments with `reply_count` greater than zero and pass their `comment_id` values into `comment_ids`: ```json { "operation": "comment_replies", "comment_ids": ["11460175497"], "max_items": 20, "max_per_target": 20 } ``` An AI assistant or automation can copy these values directly between runs. No manual ID lookup on Zhihu is needed. Keep IDs as strings, including quotation marks in JSON. This targeted operation collects replies only; `include_replies` does not affect it. Its results are saved and billed in their own run. #### Zhihu scraper input parameters · 输入参数 | Parameter | Default | Purpose | |---|---|---| | `operation` | `search` | One operation from the table above | | `keywords` | None | Search terms, up to 100 | | `max_items` | `100` | Run-wide maximum, from 1 to 10,000 | | `max_per_target` | `100` | Maximum for each keyword, URL, or comment ID, from 1 to 10,000 | | `urls` | None | Target URLs or IDs, up to 100 | | `include_replies` | `false` | Comments only: include replies as separate rows within the same result limits | | `comment_ids` | None | Top-level `comment_id` values from a Comments run | | `content_type` | `all` | Search filter or type of a bare numeric content ID | | `sort` | `relevance` | Search order: `relevance`, `upvotes`, or `newest` | | `publish_time_range` | `all` | Search period: `all`, `day`, `week`, `month`, `three_months`, `half_year`, or `year` | | `comment_sort` | `popular` | Comment order: `popular` or `newest` | | `include_html` | `true` | Include cleaned HTML alongside plain text | | `include_markdown` | `false` | Also include Markdown text | Search accepts content types `all`, `answer`, `article`, and `video`. For numeric detail IDs, select `answer`, `article`, `question`, or `pin`. Full URLs identify their own type. The Console example prefills 20 results. Programmatic input defaults to 100 when `max_items` is omitted. A **target** is one keyword, one content/profile URL, or one comment ID. Hot list uses one list. Both result limits are visible together in the first form section: `max_items` caps the whole run; `max_per_target` caps each individual input. To collect up to 500 results from one keyword, set **both limits to 500**. Increasing only the total still leaves the default 100-per-target cap in place. Limits are ceilings, not promises of availability; budget and free-plan limits also apply. ### What data can you extract from Zhihu? Content records contain an exact string ID, content type, canonical URL, title or text, available engagement counts, and author fields. Detail records can include images, topics, and video metadata. Comments and replies have `comment_id`, text, author fields, timestamps, and available counts. Replies include `parent_comment_id` so you can join them to the first-level comments. With `include_replies: true`, both row types have `operation: "comments"`; use `record_type` to distinguish them. Each reply also retains its original content ID and URL. The **Comments and replies** table view shows these links. Profiles use `user_id`, `url_token`, `name`, `profile_url`, and available biography and activity fields. Self-reported `gender` codes retain their original values (0: female, 1: male, -1: not specified), with a readable `gender_label` for recognized codes. Nothing is inferred from names or photos. Every record includes `record_type`, `operation`, and `scraped_at`. Fields that are unavailable or do not apply are omitted, not filled with nulls. Genuine zero counts and false flags are retained. Missing counts do not mean zero. Use the operation's table view for a focused export, or **All fields** to inspect the full schema. Search can return different content types in the same run; use `record_type` and `content_type` to distinguish them. The **Run summary** reports saved item counts, failed targets, skipped unreadable records, partial results, and whether an input, trial, or spending limit was reached. Good records remain saved when another record or target fails. Summary records are not billed as scraped results. #### Complete Zhihu output examples These are real rows collected on September 10, 2026. Every key and value in each selected row is included, without shortened text or arrays. Field availability varies between records. These examples are not a promise that every field appears in every operation. **Search result:** this answer matched `人工智能` in the search above. ```json { "record_type": "content", "content_id": "2063420040297308669", "content_type": "answer", "content_url": "https://www.zhihu.com/question/2042649810709239000/answer/2063420040297308669", "title": "为什么体制内至今仍不鼓励用人工智能？", "summary": "相反，体制内一开始很鼓励探索使用人工智能，特别是deepseek刚出来那会，各地政府纷纷关注，几个月之内就上线了一批本地模型", "content": "相反，体制内一开始很鼓励探索使用人工智能，特别是deepseek刚出来那会，各地政府纷纷关注，几个月之内就上线了一批本地模型。\n说到这，直接断了某些擅长政商关系的企业发财路子，之前这些企业就是准备靠个拉跨的闭源大模型，用精美PPT、宏大愿景和公关营销手段收割地方政府。\n至于为什么目前不太鼓励，其他回答都说的很到位了。", "content_html": "

相反，体制内一开始很鼓励探索使用人工智能，特别是deepseek刚出来那会，各地政府纷纷关注，几个月之内就上线了一批本地模型。

说到这，直接断了某些擅长政商关系的企业发财路子，之前这些企业就是准备靠个拉跨的闭源大模型，用精美PPT、宏大愿景和公关营销手段收割地方政府。

至于为什么目前不太鼓励，其他回答都说的很到位了。

", "author_user_id": "3532c75b53b9aa5c1327a0eeb953ab33", "author_name": "长歌当行", "author_url_token": "zai-ping-lun-wo-jiu-shi-zhu", "author_profile_url": "https://www.zhihu.com/people/zai-ping-lun-wo-jiu-shi-zhu", "author_avatar_url": "https://picx.zhimg.com/50/v2-2bebe6ea3e0e88eaaa9deb73a71b2b59_l.jpg?source=4e949a73", "author_bio": "重要吗？不重要。", "author_follower_count": 363, "question_id": "2042649810709239000", "question_url": "https://www.zhihu.com/question/2042649810709239000", "upvote_count": 2130, "comment_count": 50, "answer_count": 592, "publish_time": 1784737674, "update_time": 1784737674, "operation": "search", "scraped_at": "2026-09-10T06:25:25.190110+00:00", "query_keyword": "人工智能" } ``` **Question detail:** input `{"operation":"content_detail","urls":["21277368"],"content_type":"question","include_markdown":true}`. ```json { "record_type": "content", "content_id": "21277368", "content_type": "question", "content_url": "https://www.zhihu.com/question/21277368", "title": "如何自学人工智能？", "summary": "零基础自学人工智能，希望能推荐下学习路径。基础只有兴趣，不要说不可能。兴趣为解决人类智能缘起的问题。", "content": "零基础自学人工智能，希望能推荐下学习路径。\n\n基础只有兴趣，不要说不可能。\n\n兴趣为解决人类智能缘起的问题。", "content_html": "零基础自学人工智能，希望能推荐下学习路径。

基础只有兴趣，不要说不可能。

兴趣为解决人类智能缘起的问题。", "content_markdown": "零基础自学人工智能，希望能推荐下学习路径。 \n \n基础只有兴趣，不要说不可能。 \n \n兴趣为解决人类智能缘起的问题。", "question_id": "21277368", "question_url": "https://www.zhihu.com/question/21277368", "comment_count": 13, "view_count": 4801869, "answer_count": 632, "follower_count": 20836, "publish_time": 1372670299, "update_time": 1372670467, "operation": "content_detail", "scraped_at": "2026-09-10T06:25:37.072288+00:00", "input_url": "https://www.zhihu.com/question/21277368" } ``` **Video pin detail:** input `{"operation":"content_detail","urls":["https://www.zhihu.com/pin/2080624622454813370"],"include_markdown":true}`. Playback URLs can expire. ```json { "record_type": "content", "content_id": "2080624622454813370", "content_type": "pin", "content_url": "https://www.zhihu.com/pin/2080624622454813370", "title": "AI人工智能的福利来了！ | #人工智能 #机器人 #AI智能体 #元宇宙大模型 #自动驾驶 #我的有痛开学 #元宇宙模型 #科技前沿 #智能革命 #科学梗图 414号文件要求加速人工智能应用并大规模发展！", "content": "AI人工智能的福利来了！ | #人工智能 #机器人 #AI智能体 #元宇宙大模型 #自动驾驶 #我的有痛开学 #元宇宙模型 #科技前沿 #智能革命 #科学梗图 414号文件要求加速人工智能应用并大规模发展！", "content_html": "AI人工智能的福利来了！ |

#人工智能 #机器人 #AI智能体 #元宇宙大模型 #自动驾驶 #我的有痛开学 #元宇宙模型 #科技前沿 #智能革命 #科学梗图 414号文件要求加速人工智能应用并大规模发展！

", "content_markdown": "AI人工智能的福利来了！ |\n\n[#人工智能](https://www.zhihu.com/topic/19551275) [#机器人](https://www.zhihu.com/topic/19551273) [#AI智能体](https://www.zhihu.com/topic/28266286) [#元宇宙大模型](https://www.zhihu.com/topic/2080625086252564493) [#自动驾驶](https://www.zhihu.com/topic/19635352) [#我的有痛开学](https://www.zhihu.com/topic/2074558476941005863) [#元宇宙模型](https://www.zhihu.com/topic/29820441) [#科技前沿](https://www.zhihu.com/topic/26611258) [#智能革命](https://www.zhihu.com/topic/30694025) [#科学梗图](https://www.zhihu.com/topic/2080338645668585602) 414号文件要求加速人工智能应用并大规模发展！", "author_user_id": "d8453d3fc61ff5955ba3857cfe8e5a85", "author_name": "曾盛华ZOZOFA甑发", "author_url_token": "zozofa", "author_profile_url": "https://www.zhihu.com/people/zozofa", "author_avatar_url": "https://pica.zhimg.com/v2-aa79ea058a3fb469807b433cabe906df_720w.jpg?source=8a6f5038&needBackground=1", "author_bio": "元宇宙生态链商业模式专家、工业设计、软硬件开发、数字化转型", "author_follower_count": 36, "author_verification_label": "元宇宙生态链商业模式专家、工业设计、软硬件开发、数字化转型", "like_count": 0, "comment_count": 0, "collect_count": 0, "publish_time": 1788839676, "update_time": 1788839676, "topics": [ { "id": "2080338645668585602", "name": "科学梗图" }, { "id": "30694025", "name": "智能革命" }, { "id": "26611258", "name": "科技前沿" }, { "id": "29820441", "name": "元宇宙模型" }, { "id": "2074558476941005863", "name": "我的有痛开学" }, { "id": "19635352", "name": "自动驾驶" }, { "id": "2080625086252564493", "name": "元宇宙大模型" }, { "id": "28266286", "name": "AI智能体" }, { "id": "19551273", "name": "机器人" }, { "id": "19551275", "name": "人工智能" } ], "media_type": "video", "video_play_url": "https://vdn3.vzuu.com/HD/c06b262c-ab38-11f1-89d9-c64149588a81-v4_f2_t2_8ZQEpFLQ.mp4?auth_key=1789025137-0-0-95f21ced13ceadc18a7821dba4f8781e&bu=3a8548f7&c=avc.4.0&disable_local_cache=1&expiration=1789025137&f=mp4&pu=3a8548f7&v=tx", "video_duration_seconds": 36, "video_width": 720, "video_height": 1280, "operation": "content_detail", "scraped_at": "2026-09-10T06:25:37.072314+00:00", "input_url": "https://www.zhihu.com/pin/2080624622454813370" } ``` **Comment:** from the article-comments input above. ```json { "record_type": "comment", "comment_id": "11472791656", "content_id": "2032860336215307118", "content_type": "article", "content_url": "https://zhuanlan.zhihu.com/p/2032860336215307118", "content": "用现代模型做sft不就已经污染了。。", "content_html": "用现代模型做sft不就已经污染了。。", "author_user_id": "9a712c9b919973a61ac06162f7fe90c1", "author_name": "咸蛋", "author_url_token": "xian-dan-61-21", "author_profile_url": "https://www.zhihu.com/people/xian-dan-61-21", "author_avatar_url": "https://pic1.zhimg.com/76aef132b14be7b15951487fa6180a24_l.jpg?source=06d4cd63", "author_bio": "游戏美术设计", "like_count": 51, "reply_count": 0, "is_pinned": false, "is_author_comment": false, "publish_time": 1777453504, "operation": "comments", "scraped_at": "2026-09-10T06:25:39.693226+00:00", "input_url": "https://zhuanlan.zhihu.com/p/2032860336215307118" } ``` **Reply:** input `{"operation":"comment_replies","comment_ids":["11460175497"],"max_items":3}`. ```json { "record_type": "reply", "comment_id": "11502859456", "content_type": "answer", "parent_comment_id": "11460175497", "reply_to_comment_id": "11460175497", "content": "正常正常，心态崩的这一刻，系统性学习就算正式开始了[滑稽]", "content_html": "正常正常，心态崩的这一刻，系统性学习就算正式开始了[滑稽]", "author_user_id": "4aff31fcd26581b884c62fe3880b525b", "author_name": "啦啦啦啦", "author_url_token": "11-11-98-14", "author_profile_url": "https://www.zhihu.com/people/11-11-98-14", "author_avatar_url": "https://pic1.zhimg.com/v2-2ef8da61f9fd3376b0757aa59bee2f80_l.jpg?source=06d4cd63", "author_bio": "AI 工具评测｜已发2篇SCI 1区论文｜外企研发工作｜副业领航者", "author_verification_label": "广西大学机械硕士", "like_count": 12, "reply_count": 0, "is_pinned": false, "is_author_comment": false, "publish_time": 1780973299, "operation": "comment_replies", "scraped_at": "2026-09-10T06:25:41.160916+00:00" } ``` **Creator profile:** input `{"operation":"user_profile","urls":["11-11-98-14"]}`. ```json { "record_type": "profile", "user_id": "4aff31fcd26581b884c62fe3880b525b", "name": "啦啦啦啦", "url_token": "11-11-98-14", "profile_url": "https://www.zhihu.com/people/11-11-98-14", "avatar_url": "https://pic1.zhimg.com/v2-2ef8da61f9fd3376b0757aa59bee2f80_xl.jpg?source=32738c0c&needBackground=1", "bio": "AI 工具评测｜已发2篇SCI 1区论文｜外企研发工作｜副业领航者", "follower_count": 10044, "verification_label": "广西大学机械硕士", "gender": 0, "gender_label": "female", "answer_count": 242, "article_count": 109, "is_organization": false, "ip_region": "IP 属地广东", "employments": [ { "job": "清华大学联培", "company": "北京五道口中关村科技园" } ], "operation": "user_profile", "scraped_at": "2026-09-10T06:34:15.658237+00:00", "input_url": "https://www.zhihu.com/people/11-11-98-14" } ``` Public `ip_region` is a platform-provided region label, not a precise location or home address. Employment names are self-reported. Do not treat either as independently verified identity information. ### Zhihu API workflows and monitoring Connect keyword research, creator analysis, and comment analysis across separate runs. #### Answers: collect a question's available responses ```json { "operation": "question_answers", "urls": ["21277368"], "max_items": 20 } ``` #### Creators: collect published articles ```json { "operation": "user_content", "urls": ["https://www.zhihu.com/people/11-11-98-14"], "max_items": 20 } ``` Use `user_profile` with the same input to collect the creator's profile instead. Use `comment_replies` with `comment_ids` taken from a comments run to retrieve replies separately. #### Trends · 知乎热榜: collect the current hot list ```json { "operation": "hot_list", "max_items": 30 } ``` The number of available hot-list entries can be lower than your requested limit. ### How much does it cost to scrape Zhihu? **Saved-result charges: $2.99 per 1,000 results.** See the Actor's Pricing tab for applicable platform events and your account's current rate. One result means one saved content, comment, reply, or profile record. Duplicates within a run and failed lookups do not produce a result charge. An empty run can still incur platform charges. Free-plan access is limited to **25 results per run and 100 lifetime results**. These are access limits, not a separate promise that usage is unbilled; applicable charges use your Apify balance or credits. Free-plan runs process the **first keyword or target only** and use a bounded collection window. There is no lifetime run-count cap, and completed empty runs do not consume the result allowance. Runs may finish below the requested result limit; check the Run summary. If a run is interrupted before its saved results can be confirmed, its reserved allowance may count toward the lifetime limit. ### How to automate Zhihu research Schedule keyword searches daily with `sort: "newest"` and `publish_time_range: "day"` for topic monitoring. A six-hour hot-list schedule can capture changes in ranked discussions. Scheduling is optional; no recurring runs are enabled automatically. Use Google Sheets for review, Make or Zapier for downstream workflows, and webhooks to notify your system when a run finishes. Keep collection and analysis separate so you can retry analysis without collecting the same data again. For history, retain each dataset with its run ID and collection time. Join snapshots using string IDs and compare the same operation across dates. An engagement change is a snapshot difference, not proof of exactly when an interaction happened. **Official Apify walkthrough: set up and run an Actor** https://www.youtube.com/watch?v=1OW8gOqlZbY **Official Apify walkthrough: schedule an Actor** https://www.youtube.com/watch?v=1jI7WcVQmwM ### Zhihu scraping FAQ · 常见问题 #### Is there a Zhihu API for public search and Q\&A data? This Actor provides access through the Apify API; it is not an official Zhihu product. Supported operations and output fields are described here and in the Input tab. Private content, account actions, and unrestricted archival access are outside its scope. #### How can I get Zhihu comments from high-vote answers? Search with `content_type: "answer"` and `sort: "upvotes"`. Pass selected answer URLs to a separate `comments` run with `include_replies: true` to include reply threads. Use `comment_replies` and selected `comment_id` values when you want only specific threads. Each run saves and charges its own records. #### Can I use this Zhihu scraper from Python? Yes. Use the Apify Python client or ordinary HTTP requests to start a run and retrieve its dataset. The API tab provides language-specific examples; the AI-assistant block links to the complete input/output specification. #### What limits should I use for reliable Zhihu collection? Begin with 20 results, inspect the output, then increase the limits. Availability and runtime depend on the operation and target. Avoid overlapping schedules that collect the same topic unnecessarily; use a maximum run charge as an additional budget ceiling. #### What is Zhihu? Zhihu, written 知乎 and pronounced Zhīhū, is a Chinese question-and-answer and publishing platform. Its name roughly means “do you know?” #### Do I need a Zhihu account or cookies? No. Supply keywords, public URLs, or supported IDs. Do not include passwords or cookies in your input. #### Can I download everything from a question? Not reliably in this version. Initial answer batches work, but continuation may stop early. Check `partial` and `failed_targets` in the run summary. #### Does creator content include answers? No. `user_content` returns articles. To collect answers, use `question_answers` for a known question or `content_detail` for known answer URLs. #### Does search include full text? Search results can contain excerpts or truncated content. Run returned `content_url` values through `content_detail` when you need the available detail text. That second run produces separately billed records. #### Are replies included automatically? Not by default. Enable **Include replies · 包含回复** with the **Comments** operation to collect them automatically from content URLs. Replies are separate rows, linked by `parent_comment_id`, and count toward the same result and spending limits. If a thread cannot be completed, available rows are kept and the run summary flags partial results. The separate **Comment replies** operation remains available for collecting specific threads by ID. #### Why do counts differ between operations? Different views of a record can expose different fields or update at different times. A missing field means it was unavailable in that response. #### Why did I receive fewer results than requested? There may be fewer public matches, duplicates, unavailable targets, or collection limits. A valid empty result is different from a failed request; failures appear in the summary. #### How are duplicates handled? Records are deduplicated within each run. A restarted run can repeat records if interruption occurs between saving data and saving progress. Targets complete concurrently. When keywords overlap, `query_keyword` keeps the first saved match. Separate runs are not deduplicated against one another. #### How do I export results? Export the dataset as JSON, CSV, Excel, or HTML in Apify. JSON best preserves arrays such as images and topics. Content IDs remain strings to preserve long identifiers. Publication timestamps are Unix seconds; `scraped_at` is ISO 8601 UTC. #### Is collecting this data automatically legal? Public availability is not blanket permission. Check Zhihu's terms and the laws applicable to your use. Creator names, profiles, and comments can contain personal data; consider applicable requirements under PIPL, GDPR, and CCPA. ### Support Use the Actor's Issues tab for bugs, feature requests, or a custom collection workflow. Include the operation, affected public URL, and run summary. Never include account credentials. ### More Zen Studio scrapers for Chinese platforms **🎬 Short-video & social** -

**Weibo 微博** - [Weibo Search Scraper](https://apify.com/zen-studio/weibo-search-scraper) - [Weibo Comments Scraper](https://apify.com/zen-studio/weibo-comments-scraper) -

**Kwai (Brazil)** - [Kwai Search Scraper](https://apify.com/zen-studio/kwai-search-scraper) - [Kwai Comments Scraper](https://apify.com/zen-studio/kwai-comments-scraper) -

**RedNote 小红书** - [RedNote Search Scraper](https://apify.com/zen-studio/rednote-search-scraper) - [RedNote User Profile Scraper](https://apify.com/zen-studio/rednote-user-profile-scraper) - [RedNote Comments Scraper](https://apify.com/zen-studio/rednote-comments-scraper) - [RedNote Product Search Scraper](https://apify.com/zen-studio/rednote-product-search-scraper) - [RedNote Note Detail Scraper](https://apify.com/zen-studio/rednote-note-detail-scraper) - [RedNote Product Reviews Scraper](https://apify.com/zen-studio/rednote-product-reviews-scraper) - [RedNote Live Recorder](https://apify.com/zen-studio/rednote-live-recorder) - [RedNote Product Detail Scraper](https://apify.com/zen-studio/rednote-product-detail-scraper) -

**Douyin 抖音** - [Douyin Search Scraper](https://apify.com/zen-studio/douyin-search-scraper) - [Douyin Profile Scraper](https://apify.com/zen-studio/douyin-profile-scraper) - [Douyin Video Scraper](https://apify.com/zen-studio/douyin-video-scraper) - [Douyin Comments Scraper](https://apify.com/zen-studio/douyin-comments-scraper) - [Douyin Transcripts Scraper](https://apify.com/zen-studio/douyin-transcripts-scraper) - [Douyin Creator Rankings Scraper](https://apify.com/zen-studio/douyin-xingtu-rankings-scraper) - [Douyin Live Recorder](https://apify.com/zen-studio/douyin-live-recorder) - [Douyin Hot Search Scraper](https://apify.com/zen-studio/douyin-hot-search-scraper) - [Douyin Product Search Scraper](https://apify.com/zen-studio/douyin-product-search-scraper) -

**Xigua 西瓜视频** - [Xigua Video Search Scraper](https://apify.com/zen-studio/xigua-video-search-scraper) - [Xigua Video Detail Scraper](https://apify.com/zen-studio/xigua-video-detail-scraper) -

**Bilibili 哔哩哔哩** - [Bilibili Video Scraper](https://apify.com/zen-studio/bilibili-video-scraper) - [Bilibili Danmaku Scraper](https://apify.com/zen-studio/bilibili-danmaku-scraper) **🛒 E-commerce** -

**Naver Search 네이버 검색** - [Naver Search Scraper](https://apify.com/zen-studio/naver-search-scraper) -

**Shopee** - [Shopee Product Detail Scraper](https://apify.com/zen-studio/shopee-product-detail-scraper) - [Shopee Product Reviews Scraper](https://apify.com/zen-studio/shopee-product-reviews-scraper) - [Shopee Product Scraper](https://apify.com/zen-studio/shopee-product-scraper) -

**Taobao 淘宝** - [Taobao Search Scraper](https://apify.com/zen-studio/taobao-search-scraper) - [Taobao Product Detail Scraper](https://apify.com/zen-studio/taobao-detail-scraper) - [Taobao Seller Products Scraper](https://apify.com/zen-studio/taobao-seller-products-scraper) -

**JD.com 京东** - [JD.com Search Scraper](https://apify.com/zen-studio/jd-com-search-scraper) -

**1688 阿里巴巴** - [1688 Wholesale Scraper](https://apify.com/zen-studio/1688-wholesale-scraper) - [Alibaba Scraper](https://apify.com/zen-studio/alibaba-scraper) - [1688 Reviews Scraper](https://apify.com/zen-studio/1688-reviews-scraper) -

**Goofish 闲鱼** - [Goofish (Xianyu) Search Scraper](https://apify.com/zen-studio/goofish-xianyu-search-scraper) - [Goofish (Xianyu) Seller Scraper](https://apify.com/zen-studio/goofish-xianyu-seller-scraper) - [Goofish (Xianyu) Item Detail Scraper](https://apify.com/zen-studio/goofish-xianyu-item-detail-scraper) - [Goofish (Xianyu) Item Comments & Q\&A Scraper](https://apify.com/zen-studio/goofish-xianyu-item-comments-scraper) - [Goofish (Xianyu) Reverse Image Search Scraper](https://apify.com/zen-studio/goofish-xianyu-visual-search-scraper) -

**Coupang 쿠팡** - [Coupang Review Scraper](https://apify.com/zen-studio/coupang-review-scraper) - [Coupang Search Scraper](https://apify.com/zen-studio/coupang-search-scraper) - [Coupang Product Detail Scraper](https://apify.com/zen-studio/coupang-detail-scraper) **🏠 Real estate & autos** -

**Anjuke 安居客** - [Anjuke For-Sale Listings Scraper](https://apify.com/zen-studio/anjuke-real-estate-scraper) - [Anjuke New-Home Scraper](https://apify.com/zen-studio/anjuke-new-home-scraper) - [Anjuke Rental Listings Scraper](https://apify.com/zen-studio/anjuke-rent-scraper) - [Anjuke Listing Detail Scraper](https://apify.com/zen-studio/anjuke-sale-detail-scraper) -

**58.com 58同城** - [58.com Real Estate Scraper](https://apify.com/zen-studio/58-real-estate-scraper) -

**Autohome 汽车之家** - [Autohome (Che168) Used-Car Scraper](https://apify.com/zen-studio/che168-car-scraper) *** *Zhihu (知乎) scraper and API for keyword search, answers, articles, comments, replies, creator profiles, and trending questions.* # Actor input Schema ## `operation` (type: `string`): Choose one operation per run.

Question answers: collects an initial batch of up to 100 entries; a complete answer list is not guaranteed, and incomplete targets are flagged.
Creator articles: collects published articles, not the creator's answers. ## `keywords` (type: `array`): Search only. Add Chinese or English terms, one per entry. Up to 100 keywords.

Each keyword is searched independently, but Maximum total results is shared across them. Duplicate results are saved once. ## `max_items` (type: `integer`): Set the run-wide ceiling across all keywords or targets. For example, 20 means at most 20 rows in total, not 20 per keyword.

The example uses 20; omitted programmatic input defaults to 100. The per-target limit and available budget also apply. You are charged only for saved results, excluding duplicates and failed targets.

Free-plan access: 25 results per run and 100 lifetime results, processing the first keyword or target within a bounded collection window. No lifetime run-count cap. ## `max_per_target` (type: `integer`): Set the ceiling for each individual input: one keyword in Search, one URL for URL-based operations, or one comment ID in Comment replies. For Hot list, this limits the single list.

Maximum total results still caps the whole run. For example, 50 per keyword and 100 total across three keywords can save at most 100 rows. To collect up to 500 rows from one keyword, set both limits to 500.

These are ceilings, not guaranteed counts. Availability, budget, and free-plan limits still apply. ## `urls` (type: `array`): Add up to 100 URLs or IDs for the selected operation.

Question answers: https://www.zhihu.com/question/21277368 or a question ID.
Content details: question URLs, answer URLs such as https://www.zhihu.com/answer/2018829744922206890, article URLs such as https://zhuanlan.zhihu.com/p/2032860336215307118, or pin URLs.
Comments: answer, article, or pin URLs, not question URLs.
Creator profiles / articles: https://www.zhihu.com/people/11-11-98-14 or the profile slug 11-11-98-14.

Full URLs identify their own content type. For bare detail/comment IDs, select the matching Content type. Standalone /zvideo/ URLs are not supported; pin video URLs are. ## `include_replies` (type: `boolean`): Comments only. Also collect replies under the comments found at your answer, article, or pin URLs. No comment IDs needed.

Each reply is a separate row linked by parent\_comment\_id. Comments and replies share both result limits above and your spending limit; each saved row has the same result price. Threads are collected after their parent comments, so a small limit may stop partway through a thread.

Off by default for top-level comments only. This does not change the separate Comment replies operation. ## `comment_ids` (type: `array`): Comment replies · 评论回复 only. For the simpler URL-based workflow, select Comments and enable Include replies instead; leave this field empty.

To collect replies to specific comments, take their comment\_id values from a previous Comments dataset. Prefer top-level comments with reply\_count greater than zero. Your AI assistant or automation can pass these values straight into comment\_ids; no manual lookup on Zhihu is needed.

Use the top-level comment's ID, not an answer or article ID. Enter up to 100 IDs, one per entry, and keep them as strings in JSON to preserve their full precision. ## `content_type` (type: `string`): Search: choose All types, Answers, Articles, or Videos.

Bare numeric detail/comment IDs: choose Answers, Articles, or Pins. Questions is for question details; Question answers already knows its input type. Full URLs identify their own type.

Pins and Questions are ID-type choices, not dedicated search filters. ## `sort` (type: `string`): Search only. Choose the ranking within each keyword's results. Leave Relevance selected for a standard search.

Results from different keywords can be interleaved; this does not globally sort the combined dataset. ## `publish_time_range` (type: `string`): Search only. Restrict results to the selected publication window. Use Past week for recent discussions, or Any time for broader coverage. ## `comment_sort` (type: `string`): Comments and Comment replies only. Choose Popular for relevance or Newest for recent discussion. ## `include_html` (type: `boolean`): Keep cleaned content\_html alongside plain content where available. Enabled by default so paragraphs, links, and formatting remain usable. Turn off for smaller exports. ## `include_markdown` (type: `boolean`): Also add content\_markdown where available, useful for document processing and knowledge bases. Off by default; enable only when needed. ## Actor input object example ```json { "operation": "search", "keywords": [ "人工智能" ], "max_items": 20, "max_per_target": 100, "include_replies": false, "content_type": "all", "sort": "relevance", "publish_time_range": "all", "comment_sort": "popular", "include_html": true, "include_markdown": false } ``` # Actor output Schema ## `results` (type: `string`): Saved results for the selected operation. Choose a table view or export JSON, CSV, Excel, or JSONL. ## `summary` (type: `string`): Result totals, limits, and any targets that could not be completed. # API You can run this Actor programmatically using our API. Below are code examples in JavaScript, Python, and CLI, as well as the OpenAPI specification and MCP server setup. ## JavaScript example ```javascript import { ApifyClient } from 'apify-client'; // Initialize the ApifyClient with your Apify API token // Replace the '' with your token const client = new ApifyClient({ token: '', }); // Prepare Actor input const input = { "keywords": [ "人工智能" ], "max_items": 20 }; // Run the Actor and wait for it to finish const run = await client.actor("zen-studio/zhihu-scraper").call(input); // Fetch and print Actor results from the run's dataset (if any) console.log('Results from dataset'); console.log(`💾 Check your data here: https://console.apify.com/storage/datasets/${run.defaultDatasetId}`); const { items } = await client.dataset(run.defaultDatasetId).listItems(); items.forEach((item) => { console.dir(item); }); // 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/js/docs ``` ## Python example ```python from apify_client import ApifyClient # Initialize the ApifyClient with your Apify API token # Replace '' with your token. client = ApifyClient("") # Prepare the Actor input run_input = { "keywords": ["人工智能"], "max_items": 20, } # Run the Actor and wait for it to finish run = client.actor("zen-studio/zhihu-scraper").call(run_input=run_input) # Fetch and print Actor results from the run's dataset (if there are any) print(f"💾 Check your data here: https://console.apify.com/storage/datasets/{run.default_dataset_id}") for item in client.dataset(run.default_dataset_id).iterate_items(): print(item) # 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/python/docs/quick-start ``` ## CLI example ```bash echo '{ "keywords": [ "人工智能" ], "max_items": 20 }' | apify call zen-studio/zhihu-scraper --silent --output-dataset ``` ## MCP server setup ```json { "mcpServers": { "apify": { "type": "http", "url": "https://mcp.apify.com/?tools=fetch-actor-details,zen-studio/zhihu-scraper" } } } ``` The hosted server signs you in with OAuth on first connect, so no API token belongs in this config. Clients without OAuth support can send an `Authorization: Bearer ` header instead, using a token from API & Integrations in Apify Console (https://console.apify.com/settings/integrations). ## OpenAPI specification Download the OpenAPI definition: https://api.apify.com/v2/actors/YOq90GHMVnTe8qtHO/builds/7OQ8oxQ3tp42gOWhH/openapi.json