RedNote (Xiaohongshu) Search Scraper

What is RedNote (Xiaohongshu) Search Scraper?

RedNote (Xiaohongshu) Search Scraper is a Xiaohongshu scraper that finds public RedNote notes by keyword, note link, creator link or explore category and returns one structured row per note: title, full text, hashtags, likes, collects, comments, shares, author, publish time, images and video links. It is built for marketers, brand teams, influencer researchers and analysts: just enter a keyword such as 咖啡 or lululemon and click Start. No RedNote account, login or cookies are needed, and the Apify Free plan lets you save up to 10 results per run to try it.

Whether you call it a RedNote scraper, an XHS scraper, a Little Red Book scraper or a Xiaohongshu crawler, this RedNote Xiaohongshu scraper does the same job: it turns public Xiaohongshu notes into clean rows you can filter, sort, export and analyse.

• Xiaohongshu marketing research: see which notes about your product category earn the most likes, collects and shares.
• Xiaohongshu brand monitoring: run your brand name as a keyword on a schedule and follow what people post about it.
• RedNote influencer marketing: find the authors of high-engagement notes, and turn on author details to get follower counts, location and bio.
• Hashtag research: collect RedNote hashtags for a keyword and see the top hashtags per keyword in the summary table.
• Follow Xiaohongshu trends: combine the time window with sorting by likes or newest to see what is getting attention now.
• Xiaohongshu data collection for analysis and AI: every row is structured JSON, ready for spreadsheets, BI tools, sentiment analysis or retrieval pipelines.

What can RedNote (Xiaohongshu) Search Scraper do?

• 🔍 Search notes in bulk by keyword (Chinese or English), note link, xhslink.com share link, note ID, creator link or RedNote search page link, all in one list. It works as a Xiaohongshu search scraper and a Xiaohongshu note scraper at the same time.
• 🧭 Collect explore categories such as food, travel, fashion or 美食, or the main recommendation feed.
• 🎯 Filter every note by note type (image or video), time window, publish dates and minimum likes, collects and comments, and check that the keyword really appears in the note.
• 🏆 Sort results by relevance (as found), likes, newest, collects, comments, shares or total engagement.
• 📸 Add media and author details: original-quality image links, video links with chapters, covers, live-photo clips and optional author profiles.
• 📊 Get a keyword summary: median likes, collects and comments, video share, top hashtags, top notes and top creators per keyword.
• 📦 Export to JSON, CSV, Excel, XML or HTML, or use the Apify API, schedules, webhooks and integrations.
• 🛡️ Starts without a proxy and switches to Apify Proxy automatically when requests are refused.

---

Xiaohongshu scraper: what data can it extract?

Each note becomes one row of public Xiaohongshu data. Together, the fields give you RedNote data you can analyse right away, from engagement numbers to Xiaohongshu hashtags and author profiles. Field names below are the exact names in the dataset.

🔗 Note ID and link: noteId, url	🎬 Note type: noteType (image or video)
📝 Title and full text: title, description	#️⃣ Hashtags: hashtags (without #)
👤 Mentioned accounts: mentions	📅 Published and last updated: publishedAt, lastUpdatedAt
📍 IP location label: ipLocation	🏷️ Author statement (for example AI-generated content): authorStatement
❤️ Likes: likes	⭐ Collects (saves): collects
💬 Comments (count): comments	🔁 Shares: shares
📈 Total engagement: engagementTotal	➗ Collect-to-like ratio: collectToLikeRatio
≈ Rounded counts flag: countsApproximate	👤 Author: author (userId, nickname, avatar, profileUrl)
👥 Author details: authorDetails (optional)	📸 Cover, images and video: coverUrl, images, video
🔎 Keyword and match: keyword, keywordMatchedIn	🧭 How it was found: source, sourceRank
🕒 Collection time: scrapedAt

Note: RedNote doesn't show every value on every note. lastUpdatedAt, ipLocation and authorStatement are null when the note doesn't show them. Counts the site rounds (such as 1.2万) are converted to numbers and flagged with countsApproximate: true.

Two more tables are part of the output: Creators (filled when you add creator links: one row per creator link, with followers, following, likes and collects, notes posted, verification, tags and recent note cards) and Keyword summary (one row per keyword that saved notes, plus rows for your note links and categories).

What data can't it extract?

• Comment text and replies (you get the comment count only)
• Private notes, or anything that needs a logged-in RedNote account
• Follower or following lists
• Links to a creator's recent notes (creator rows show the title, type, likes, pinned flag, date and cover of each recent note card, without a note link)
• Translations: notes are returned in the language they were posted in

---

How to scrape Xiaohongshu notes

This RedNote search scraper is designed for an easy start, even if you have never extracted data from the web before. You don't need to write code to scrape Xiaohongshu.

1. Create a free Apify account using your email.
2. Open RedNote (Xiaohongshu) Search Scraper.
3. In 🔗 Keywords and RedNote links (bulk), enter one keyword or link per line. The form starts with 咖啡 (coffee).
4. Click Start and watch the log. With the default sort, notes appear in the Output tab as soon as they are found.
5. Download your data in JSON, CSV, Excel, XML or HTML.

⚠️ Your first run will look small on purpose. 📊 Max results per keyword starts at 10, and the Free plan saves up to 10 results per run in total. Raise the limit (on a paid plan) when the test looks right.

Fewer than 10 rows is normal too: 🎯 Keyword must appear in the note drops off-topic notes, and some recent notes are skipped (see Why am I getting fewer results than expected?). Creator links are processed last, after note links, keywords and categories, so on the Free plan test creator links in their own run.

---

How much does it cost to scrape Xiaohongshu?

The Actor uses pay-per-event pricing: you pay a fixed price for each saved note or creator row, plus a small fee each time a run starts. Your bill has three parts:

Part	Price	When it is charged
💲 Results	$3.00 per 1,000 results ($0.003 per row) on the Free plan (lower on paid plans, see below)	For each note row and each creator row saved
🚀 Actor start	$0.00005 per run at the default memory (1 GB)	Once per run
⚙️ Platform usage	Apify's normal usage rates for your plan	On top of the two event prices (see below)

Apify plan	Price per result	Price per 1,000 results
Free	$0.003	$3.00
Starter	$0.0027	$2.70
Scale	$0.0025	$2.50
Business and higher	$0.002	$2.00

Paid Apify plans get a lower price per result. The plan discount applies to that price only. The start fee is the same on every plan, and platform usage is billed at your Apify plan's own rates.

• Not charged as results: keyword summary rows, the run summary record, and notes that were skipped or removed by your filters.
• No double charges for duplicates: a note is saved once per run, even when several keywords find it.

Rounded examples for the result and start charges at the Free plan price, before platform usage. A single Free plan run can save up to 10 results, so on the Free plan only the 10-row example applies.

Rows saved	Results	Start fee	Event total
10	$0.03	$0.00005	≈ $0.03
100	$0.30	$0.00005	≈ $0.30
1,000	$3.00	$0.00005	≈ $3.00

What platform usage is made of:

• Compute: the run's memory (1 GB by default) multiplied by how long the run takes. Turning on author details adds one request per unique author, so the run takes longer.
• Storage and data transfer: the rows themselves. Images and videos are saved as links, not as files.
• Apify Proxy traffic: only when the run has to switch to Apify Proxy, or when you choose a proxy yourself.

How to estimate platform usage: we have no measured figures to quote, so measure it with a small test run. On the Free plan, a 10-note test costs $0.03 in result charges plus the $0.00005 start fee, plus its own platform usage:

1. Run one of your keywords with 📊 Max results per keyword set to 10 and your usual options.
2. Open the finished run and note the platform usage cost and the number of rows saved.
3. Divide the usage cost by the rows saved, multiply by the rows you plan to collect, and add the result charges and the start fee.

The run page also shows how long the test took. We don't quote run times: they grow with the number of notes, and turning on author details makes runs longer.

Keep costs under control: set a maximum charge limit when you start a run. When the limit is reached, the Actor stops cleanly and keeps everything saved so far. See the pricing tab for the current event list.

Is scraping Xiaohongshu free?

You can try it for free. The Apify Free plan gives you $5 of credit every month, and that one credit pays for the result charges and the platform usage together. Free plan runs save up to 10 results per run, counting note rows and creator rows together. A full 10-result run costs $0.03 in result charges plus the $0.00005 start fee and its platform usage. To save more than 10 results in one run, use a paid Apify plan.

---

Xiaohongshu scraper input

Only one field is required: 🔗 Keywords and RedNote links (bulk) (urls), with at least one entry. Everything else has a working default. You can fill in the form or send the same settings as JSON through the API.

Settings most people change

Form field	JSON name	Default	What it does
🔗 Keywords and RedNote links (bulk)	urls	咖啡 (prefill)	Keywords, note links, share links, note IDs, creator links and search page links, one per line
📊 Max results per keyword	maxResults	10	Notes kept per keyword after your filters. 0 = as many as can be found. It applies to keywords only: note links and creator links are all processed, up to the Free plan cap and your maximum charge limit
🧭 Explore categories	channels	empty	Categories such as food, travel, fashion or 美食. recommend = the main recommendation feed. Unknown names are skipped (see the category note below)
📚 Max notes per category	maxNotesPerChannel	10	Notes kept per category. 0 = no cap
🏆 Sort by	sortType	Most relevant (as found)	Also: most popular (likes), newest first, most collected, most commented, most shared, highest total engagement
🎬 Note type	noteType	All types	Image notes or video notes only
🕒 Time window	timeFilter	Any time	Last 24 hours, last week (7 days), last month (30 days), last 6 months (182 days) or last year (365 days)
📅 Published after / 📆 Published before	publishedAfter, publishedBefore	empty	Dates in YYYY-MM-DD. A "before" date includes that whole day
💖 Minimum likes / ⭐ collects / 💬 comments	minLikes, minCollects, minComments	0 (off)	Keep only notes with at least this many
🎯 Keyword must appear in the note	strictKeywordMatch	on	Drop keyword results whose text doesn't contain the keyword
🔤 Where to look for the keyword	keywordMatchIn	title, description, hashtags	Which parts of the note the keyword check reads
🌄 Include images and video links	includeMedia	on	Adds coverUrl, images and video
🎨 Include author details	includeAuthorDetails	off	Adds authorDetails (followers, location, bio, gender and more)

Here's an example that collects up to 10 notes for 咖啡, the same as the form's starting values:

{
  "urls": ["咖啡"],
  "maxResults": 10
}

See the input tab for a full description of all fields.

What can you put in the keywords and links box?

The box accepts plain keywords (each one is a separate search with its own result budget), note links, xhslink.com share links, bare note IDs, creator links and RedNote search page links (the keyword is read from the link). Entries it doesn't recognise are skipped, with a warning in the log.

• ✅ Good: 咖啡, lululemon, a note link copied from the RedNote website or app, a creator link copied from the site (for example from a note's author name)
• ❌ Bad: a creator link typed by hand, or a creator link shortened to just the profile ID

⚠️ Creator links must be copied in full from the website. The full address carries a code (an access token) that lets logged-out visitors open the page, and creator links without it are skipped. For the same reason, some recent notes can't be opened from a bare note ID or from keyword results; paste the full note link copied from the site to include them.

How to copy a creator link:

1. On xiaohongshu.com, open any note by the creator.
2. Click the creator's name to open their page.
3. Copy the whole address from your browser's address bar (it is long) and paste it as one line.

App share links (xhslink.com) work for notes only. A creator's app share link is skipped, so use the website link from the steps above.

Explore categories: use a category name as RedNote's explore page shows it (for example 美食), or an English name such as food, travel or fashion. Use recommend for the main recommendation feed. A name that doesn't match a category is skipped, with a warning in the log.

Sort by

Sorting is applied to the notes this run found. It is not RedNote's in-app ranking. With Most relevant (as found), every note is saved the moment it passes your filters. With any other option, rows appear after each keyword, link batch or category is finished, already sorted.

Keyword must appear in the note

This check is on by default and makes keyword results stricter. A note found for a keyword is kept only if the keyword text appears (ignoring upper and lower case) in the parts chosen in 🔤 Where to look for the keyword. It doesn't apply to note links, categories or creator links.

• ✅ Good: a brand name or a Chinese keyword that people write in their notes, such as 咖啡
• ❌ Bad: an English keyword for a topic people write about in Chinese, with the check on

⚠️ If you get few results, turn the check off to keep everything found for the keyword.

Advanced settings

Form field	JSON name	Default	What it does
🎥 Include every video quality	includeAllVideoStreams	off	Adds video.streams with every available codec and resolution
🚀 Parallel requests	maxConcurrency	4	How many notes are fetched at the same time (1-10). Lower it if you see many blocks
🐢 Pause between requests	requestDelaySeconds	1	Base pause in seconds between pages (0-30)
🔁 Retries per request	maxRetries	3	Total attempts per request, including the first one (1-10)
🌐 Proxy	proxyConfiguration	no proxy	The run starts without a proxy. When requests are refused, it switches to Apify Proxy automatically (if your account can use it) and stays on it for the rest of the run. Proxy traffic counts as platform usage. If Apify Proxy isn't available on your account, the run continues without a proxy, and notes that can't be opened are skipped. Your plan's proxy access is listed on the Apify pricing page. You can also choose a proxy yourself

---

Xiaohongshu scraper output

Results are stored in datasets that you open from the Output tab. Note rows come with six ready-made views: Overview, Engagement, Content, Authors, Images and video and Search info. The Creators and Keyword summary tables and the Run summary are there too. Each run builds a Xiaohongshu dataset you can view as a table or JSON, trim to the fields you need, and download in JSON, CSV, Excel, XML or HTML. To export later, go to Storage → Datasets, open the dataset and click Export.

JSON example

One real note row for the input example above (keyword 咖啡), shortened: the description is cut with "…", media links are replaced with "…", and source is left out.

{
  "noteId": "68d3c0a9000000001302a1ca",
  "url": "https://www.xiaohongshu.com/explore/68d3c0a9000000001302a1ca",
  "noteType": "image",
  "title": "热门冻干咖啡超全测评！打工人速看",
  "description": "打工人早8，急需快速来杯提神好咖啡…",
  "hashtags": ["三顿半", "冻干咖啡", "咖啡"],
  "mentions": [],
  "publishedAt": "2025-09-24T10:18:02Z",
  "lastUpdatedAt": "2025-10-17T04:34:07Z",
  "ipLocation": "上海",
  "authorStatement": null,
  "likes": 192,
  "collects": 67,
  "comments": 14,
  "shares": 4,
  "engagementTotal": 277,
  "collectToLikeRatio": 0.349,
  "countsApproximate": false,
  "author": {
    "userId": "5d0af5120000000016034bfd",
    "nickname": "毛毛测评记（真话版）",
    "avatar": "…",
    "profileUrl": "…"
  },
  "coverUrl": "…",
  "images": [
    { "url": "…", "width": 1920, "height": 2560, "livePhoto": false }
  ],
  "video": null,
  "keyword": "咖啡",
  "keywordMatchedIn": ["title", "description", "hashtags"],
  "sourceRank": 1,
  "scrapedAt": "2026-10-10T19:58:54Z"
}

engagementTotal is likes + collects + comments + shares (192 + 67 + 14 + 4 = 277), and collectToLikeRatio is collects divided by likes (67 / 192 ≈ 0.349).

When optional fields are filled

Field	When it has a value
coverUrl, images, video	Only when 🌄 Include images and video links is on (default). video is null for photo notes. If original-quality images can't be reached, images.url falls back to the preview link, which expires
video.streams	Only when 🎥 Include every video quality is on
authorDetails	Only when 🎨 Include author details is on. It is null when the author's page can't be read
keyword, keywordMatchedIn	For notes found by keyword. null for note links and categories
collectToLikeRatio	null when a note has 0 likes or no like count
lastUpdatedAt, ipLocation, authorStatement	When RedNote shows them on the note

Notes or creators that can't be read are skipped. The Actor doesn't write error rows, so every note row and every creator row was read successfully. The Run summary counts notes skipped because they need a link copied from the site, because they no longer exist, because the keyword wasn't in their text, or because your filters removed them.

Creators, keyword summary and run summary

• Creators: one row per creator link: userId, profileUrl, redId, nickname, gender, ipLocation, bio, avatar, followersText, followers, followersApproximate, following, likesAndCollects, notesPosted, professionTags, location, college, infoTags, verified, verifyType, recentNotes and scrapedAt. recentNotes holds as many recent note cards as the creator's page shows to logged-out visitors; the Actor adds no limit of its own.
• Keyword summary: one row per keyword, plus one row for all your note links together and one row per explore category. A keyword, link batch or category that saves no notes gets no summary row. Each row has notes, videoShare, medianLikes, medianCollects, medianComments, topHashtags (up to 15), topNotes (the 5 most-liked notes), topCreators (the 5 creators with the most likes), newestPublishedAt, oldestPublishedAt and scrapedAt. It is ready-made Xiaohongshu analytics: you can read the popular Xiaohongshu hashtags for a keyword without building a pivot table.
• Run summary: totals for the whole run, including the skipped-note counts described above.

---

Tips and advanced usage

How many results can you scrape?

You get up to 📊 Max results per keyword notes per keyword and up to 📚 Max notes per category notes per category. You can raise these up to 100,000, or set 0 for as many as can be found; on the Free plan, a run still saves up to 10 results. What you actually get also depends on how many public notes can be found for each keyword, on your filters, and on your maximum charge limit. To find out for your keyword, run a quick test.

Inputs are processed in this order: note links first, then keywords, then categories, then creator links. On the Free plan, the 10-result cap can be used up before the later inputs start.

How to get faster or cheaper results

• Keep 🎨 Include author details off unless you need follower numbers. It adds one request per unique author.
• Set 📊 Max results per keyword to what you really need, and add filters (likes, dates, note type) so you only pay for rows you want.
• If you see many blocks in the log, lower 🚀 Parallel requests or raise 🐢 Pause between requests.

Track brands and Xiaohongshu trends on a schedule

Create an Apify schedule with your brand or topic keywords, 🕒 Time window set to the last week and 🏆 Sort by set to newest first. Each run then gives you a fresh weekly list, and the keyword summary shows how median likes and top hashtags change over time.

Known issues

• Creator links without the access token part are skipped. Copy the full link from the site.
• Some recent notes found by keyword are skipped because the site opens them only from a link copied from the site. The Run summary tells you how many; paste those note links to include them.
• If no note passes your filters after many checks, the Actor stops that keyword and the log names the filter that rejected the most notes. Loosen that filter.
• If original-quality images can't be reached, preview image links are used, and they expire. Download images you need soon after the run.
• With a non-default sort, rows appear only after each keyword, link batch or category is finished.
• If RedNote changes its website, some fields may come back empty until the Actor is updated. Please open an issue if you see this.

---

Integrations and API

Can I use RedNote (Xiaohongshu) Search Scraper as an API?

Yes. The Apify API lets you use this Actor as a RedNote API for public notes: start runs, schedule them and fetch the results as JSON. If you were looking for a Xiaohongshu search API or a RedNote search API, send your keywords as input and read the notes from the dataset. This works as a Xiaohongshu scraper API for any language that can call HTTP. Use the apify-client NPM package for Node.js or the apify-client PyPI package for Python. The API tab has ready-to-use code.

Xiaohongshu Python example

from apify_client import ApifyClient

client = ApifyClient("<YOUR_API_TOKEN>")
run = client.actor("insights_data/rednote-xiaohongshu-search-scraper").call(
    run_input={"urls": ["咖啡"], "maxResults": 10}
)

for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["title"], item["likes"], item["url"])

Can I integrate it with other apps?

Yes. Connect the Actor with almost any cloud service through Apify integrations: Make, Zapier, Slack, Google Sheets, Google Drive, Airbyte, GitHub, LangChain and more. Use webhooks to trigger an action when a run finishes. For example, send every morning's notes for your brand keyword to Google Sheets and build RedNote analytics charts on top.

Can I use it with AI agents (MCP)?

Yes. If you were searching for a Xiaohongshu MCP option, connect this Actor to Claude and other AI assistants through the Apify MCP server. Your agent can then search public RedNote notes and read the results. Open the MCP tab for setup steps.

---

Other scrapers for Chinese social media

Need a Chinese social media scraper for another platform? These Actors from the same author cover more China social media data and short-video content:

🔍 Weibo Scraper — Posts, Comments, Profiles & Hot Search: keyword and date-range search, topics, hot search boards, posts, comments, profiles and KOL discovery. No account or cookie needed	🎬 Bilibili Scraper — Videos, Comments, Danmaku & Creator Uploads: videos, comments with replies, danmaku, creator uploads, profiles, rankings and trending searches
👤 TikTok Profile Scraper: videos, reposts, stories and playlists of any TikTok profile with views, likes, comments, shares, hashtags and author stats	💬 TikTok Comments & Replies Scraper: comments and replies from videos, photo posts and whole profiles, with likes, reply counts, dates and commenter details

---

FAQ

Is it legal to scrape Xiaohongshu?

Web scraping is generally legal if you extract publicly available data, and this Actor reads only notes that are visible without logging in. However, notes can contain personal data, so respect privacy laws (for example GDPR, or China's PIPL), collect personal data only with a legitimate reason, and take RedNote's terms of service into account. This is not legal advice; read Apify's guide to the legality of web scraping and ask a lawyer if you are unsure.

How does RedNote (Xiaohongshu) Search Scraper work?

It is web scraping Xiaohongshu the way a logged-out visitor browses the site, only faster. For each keyword, the Actor finds public notes that match it, opens each note, reads the text, numbers, author and media, checks your filters and saves one row. Note links are opened directly, categories are read from RedNote's explore feed, and creator links give one creator row. At the end it writes the keyword summary and the run totals.

Can I do a Xiaohongshu search without login?

Yes. You don't need a RedNote account, a login or cookies; the Actor only reads public notes. That makes it a practical way to run a RedNote search without login and get the results as data. Two things still need links copied from the site: creator links, and some recent notes found by keyword (see Known issues).

Is there an official Xiaohongshu API?

In our research we didn't find an official, open Xiaohongshu API for public note data; the API offers we found came from third-party services. This Actor gives you API access through Apify instead: you send keywords or links and get notes back as JSON (see Integrations and API).

How is it different from copying notes by hand?

	RedNote (Xiaohongshu) Search Scraper	Copying by hand
Many keywords and links at once	✅ One list, one run	❌ One search at a time
Likes, collects, comments, shares as numbers	✅ Rounded counts converted to numbers and flagged	❌ Rounded values like 1.2万
Filters and sorting	✅ Type, dates, minimum engagement, keyword check	❌ Manual
Export and scheduling	✅ JSON, CSV, Excel, API, schedules	❌ Copy-paste
Cost	$3.00 per 1,000 results on the Free plan, down to $2.00 on Business, + start fee + platform usage	Your time

Why scrape RedNote notes?

• RedNote data analysis: measure what content earns likes and collects in your niche.
• Xiaohongshu influencer marketing: shortlist creators by the engagement of their notes and their follower numbers.
• Product and brand feedback: read what people write about your brand and competitors.
• Content planning: find the titles, hashtags and formats (image or video) that work for a keyword.
• Datasets for AI: feed clean JSON into sentiment analysis, summarisation or retrieval.

Why am I getting fewer results than expected?

1. Check 📊 Max results per keyword: it starts at 10.
2. On the Free plan, a run saves up to 10 results in total.
3. Turn off 🎯 Keyword must appear in the note or add more parts in 🔤 Where to look for the keyword.
4. Loosen filters such as minimum likes, note type or dates. The log names the filter that rejected the most notes.
5. Check the Run summary for notes skipped because they need a link copied from the site, and paste those note links.
6. Add more keywords: each keyword has its own result budget, and a note found by two keywords is saved once.
7. If the run status says no notes matched, try a different keyword or check the spelling.

Can I get comments with this scraper?

You get the number of comments for each note (comments), not the comment text or replies.

---

Support and feedback

Found a bug or have a feature request? Please create an issue on the Issues tab. Your feedback helps us improve RedNote (Xiaohongshu) Search Scraper.

Need a custom solution, another field or a different data source? Contact us at hello.insights.data@gmail.com.

Built and maintained by Insights Data.