Weibo Scraper

🔴 Weibo Scraper

Search public Weibo posts by keyword, open a single post, or read a user profile, and get clean rows with the text, counts, timestamps, pictures, and links. No Weibo account or login needed. Chinese and English keywords both work, at $22 per 1,000 results.

✨ Why use Weibo Scraper?

• 🔓 No login or cookies — reads public posts and profiles only.
• 🔎 Three modes in one actor — keyword search, post detail, and user profile.
• 🈶 Chinese and English search — mixed keywords work too.
• 📅 Date range — limit a search to a start and end day in China time.
• 🖼️ Media links — picture URLs and video links are included on posts.
• ✅ Fair billing — a missing post or user saves no row and is not charged; it is listed in failedInputs of the OUTPUT record.
• 📦 Export anywhere — JSON, CSV, Excel, or the API.

🎯 Use cases

• 📈 Brand and topic monitoring — follow what people on Weibo say about a product or event.
• 🛍️ China market research — gauge interest in brands, launches, and trends.
• 🧑‍🤝‍🧑 Influencer vetting — read a profile's followers, posts count, and verification.
• 🗞️ Media tracking — collect public posts on breaking topics.
• 🧠 Text datasets — build Chinese social text corpora with timestamps and counts.

🚀 How to use

1. Pick What to scrape: keyword search, post detail, or user profile.
2. Fill the matching field: keyword, post ID or URL, or user ID or profile URL.
3. For a search, set Maximum posts and an optional date range.
4. Click Start and wait for the run to finish.
5. Export the dataset as JSON, CSV, or Excel.

📥 Input

{
  "operation": "searchWeibo",
  "keyword": "Tesla",
  "maxResults": 10,
  "startDay": "2026-09-14",
  "endDay": "2026-10-14",
  "proxyConfiguration": {
    "useScrapelyProxy": true,
    "scrapelyProxyGroups": ["DATACENTER"]
  }
}

Post detail and profile examples:

{ "operation": "weiboDetail", "weiboId": "5249718969175092" }

{ "operation": "userDetail", "userId": "2803301701" }

📤 Output

One dataset row per post or profile.

Profile rows add followersCount, friendsCount, statusesCount, description, location, verifiedReason, and avatarUrl.

{
  "weiboId": "5353827684846227",
  "bid": "RmAiQkkIX",
  "userId": "1234567890",
  "screenName": "ExampleUser",
  "text": "tesla",
  "repostCount": 0,
  "commentCount": 0,
  "attitudeCount": 0,
  "postedAt": "2026-10-14 07:50:52",
  "source": "iPhone客户端",
  "regionName": "发布于 湖北",
  "pics": [
    "https://wx1.sinaimg.cn/large/008ygTaAgy1ii1jq044ivj30zo2564m9.jpg",
    "https://wx3.sinaimg.cn/large/008ygTaAgy1ii1jq0o5yfj30zo256h2m.jpg"
  ],
  "picCount": 14,
  "videoUrl": "",
  "weiboPageUrl": "https://m.weibo.cn/detail/5353827684846227",
  "userPageUrl": "https://weibo.com/u/1234567890",
  "verified": false,
  "status": "success"
}

(Trimmed to one post with the picture list shortened. Rows also repeat a few fields under alternate names, such as screen_name and reposts_count.)

💰 Pricing

You pay $0.022 per saved post or profile, which is $22 per 1,000 results. Missing posts and users are free. Platform usage (compute and proxy) is billed on top.

Worked example: a 10-post search costs 10 × $0.022 = $0.22 in events plus about $0.003 of platform usage, around $0.223 in total.

❓ FAQ

Is it legal to scrape Weibo?

The actor reads only public posts and profiles. You are responsible for using the data in line with Weibo's terms and applicable law, especially for personal data.

Do I need a Weibo account?

No. Only public data is collected and the actor does not take your login.

Can I get follower or following lists?

No. Weibo returns those lists only to logged-in accounts.

Can I search in Chinese?

Yes. Chinese, English, and mixed keywords all work.

What if a later page fails?

Rows already saved stay in the dataset, and the run summary says the result is partial.

Can I schedule it or use the API?

Yes, from the console or through the API. If you are moving from another Weibo scraper, operation, keyword, weiboId, userId, startDay, and endDay use the same names.

⚠️ Limits

• At most 1,000 posts per search; Weibo itself stops serving a search after about 50 pages (roughly 1,000 posts). The status message says when a search ran out.
• One post or one profile per detail run, and one operation per run.
• Private accounts and deleted posts save no row; they are listed in failedInputs of the OUTPUT record.
• Search text is the public short text. Use Post detail for the expanded long text.
• Follower and following lists are not available.
• A 10-post search takes about 9 seconds including startup.