Weibo Scraper in English

Read what China says on Weibo, in English. Type a keyword in English and get Weibo posts or videos with an English translation, sentiment, the poster's province, author followers and the top comments.

No Weibo account, no Chinese and no proxy needed.

How it works

1. Type your keyword in English, for example Xiaomi car delivery. The Actor turns it into the words Chinese users actually write, 小米汽车 交付, and shows the Chinese keyword in the run status. Chinese keywords work too.
2. Choose posts or videos, how many, latest or most popular, and an optional date range.
3. Get every post with the original Chinese text and the English translation side by side, plus sentiment.

Input

Field	What it does	Default
Search query	Keyword in English or Chinese. Separate words with a space to find posts that contain all of them.	required
Search type	Posts (text and image posts) or Videos	Posts
Max posts	How many posts to collect, up to 10,000	100
Sort by	Latest or Most popular	Latest
Start date / End date	Optional date range, in China time. A date like 2026-03-01, or a relative value like 7 days	latest posts
Translate Chinese text to English	Adds textEnglish next to every Chinese text	on

Example:

{
  "query": "Xiaomi car delivery",
  "searchType": "posts",
  "maxPosts": 500,
  "sortBy": "latest",
  "startDate": "2026-03-01",
  "endDate": "2026-03-31"
}

Daily monitoring: set both Start date and End date to 1 day and schedule the Actor every morning. Each run collects all of yesterday's posts, with no duplicates between runs.

What you get

Every row is one post. type is post for text and image posts and video for video posts.

Group	Fields
Post	postId, postUrl, createdAt, timestamp, text, textEnglish
Engagement	repostsCount, commentsCount, likesCount
Context	hashtags, mentions, links (short link and its real destination), device, province
Media	imageUrls, imageCount, videoUrl, videoThumbnailUrl, videoViewCount
Author	name, profileUrl, avatarUrl, isVerified, followersCount
Top comments	Up to 5 most liked comments, each with textEnglish, province and sentiment
Sentiment	sentiment (positive, neutral, negative) and sentimentScore from -1 to 1
Repost	repostOf holds the original post when the row is a repost

Fields that do not apply are null, for example video fields on a text post.

In textEnglish, hashtags and Weibo emoji codes such as [哈哈] are translated too. @mentions stay as written, because they are the real account names you can look up on Weibo.

Output example

{
  "type": "post",
  "postId": "5279912345678901",
  "postUrl": "https://weibo.com/7654321098/RlWqErTyU",
  "createdAt": "2026-03-18T20:41:00+08:00",
  "timestamp": 1773837660,
  "text": "@小米汽车 订了YU7快八个月了还没交车，客服每次都说再等等，到底什么时候能交？#小米YU7交付#",
  "textEnglish": "@小米汽车 I ordered the YU7 almost eight months ago and still no car. Customer service keeps telling me to wait. When will it actually be delivered? #XiaomiYU7Delivery#",
  "repostsCount": 37,
  "commentsCount": 214,
  "likesCount": 1893,
  "hashtags": ["小米YU7交付"],
  "mentions": ["小米汽车"],
  "links": [],
  "device": "Xiaomi 15 Pro",
  "province": "江苏",
  "imageUrls": ["https://wx3.sinaimg.cn/large/example.jpg"],
  "imageCount": 1,
  "videoUrl": null,
  "videoThumbnailUrl": null,
  "videoViewCount": null,
  "repostOf": null,
  "author": {
    "userId": "7654321098",
    "name": "等车的老王",
    "profileUrl": "https://weibo.com/u/7654321098",
    "avatarUrl": "https://tvax3.sinaimg.cn/example.jpg",
    "isVerified": false,
    "followersCount": 326
  },
  "topComments": [
    {
      "commentId": "5279919999000001",
      "createdAt": "2026-03-18T21:05:00+08:00",
      "text": "我也等了七个月，太难了",
      "textEnglish": "I've been waiting seven months too, it's so hard",
      "likesCount": 412,
      "province": "江苏",
      "author": { "userId": "8765432109", "name": "YU7车主小陈" },
      "sentiment": "negative",
      "sentimentScore": -0.7
    }
  ],
  "sentiment": "negative",
  "sentimentScore": -0.8
}

Download results as JSON, CSV or Excel, or read them through the Apify API.

Who uses it

• Brands and PR teams watching their name, campaigns and possible crises in China.
• Agencies finding Chinese creators and checking what people say about them.
• Investors and analysts reading consumer sentiment about Chinese brands and products.
• Researchers and AI teams collecting Chinese posts with English translations.

Tips

• Date ranges are spread evenly. Ask for 500 posts in March and you get posts from every day of March, not only March 31.
• More than 500 posts is fine. Weibo shows at most 50 pages per search, so the Actor splits busy days into hours.
• Check the Chinese keyword in the run status. If it is not what you meant, type the Chinese words yourself.
• Video links expire. Weibo video links stop working after a few hours, so download videos right after the run.
• Ads and tag spam are removed automatically. Tag spam is the same text posted by many accounts that tag dozens of brands, where your keyword only appears as one of the tags. Normal posts that tag a brand, such as complaints to @小米汽车, are kept.

Notes

This Actor collects publicly available Weibo data only. Translations are made by AI and can miss slang or irony, so the original Chinese text is always included. You are responsible for using the data in line with applicable laws, including data protection laws, and Weibo's terms.