Bilibili Search Extractor

Turn any Bilibili search into clean, structured data. Give the actor a list of keywords in Chinese or English and it returns every matching video with its title, direct link, cover image, uploader, play count, danmaku count, duration, and publish date.

Bilibili is one of the largest video platforms in China, and its search results are a rich source of signals for content, marketing, and SEO research. This actor collects those results for you at scale, with no login, no browser automation to babysit, and no manual copy and paste.

Why use this actor

• Keyword research at scale. Feed a whole list of terms and get a combined dataset back in one run.
• SEO and content strategy. See which videos rank for a keyword, who publishes them, and how they perform.
• Competitor and brand monitoring. Track how a game, show, product, or campaign is covered on Bilibili over time.
• Trend spotting. Pull play and danmaku counts to find what is gaining traction in a niche.
• Ready for pipelines. Export to JSON, CSV, or Excel, or pull results straight from the dataset API.

Features

• Search Bilibili by any keyword, in Chinese or English
• Batch mode: pass many keywords in a single run
• Automatic pagination up to the limit you set
• Per keyword limit, so each term gets a fair share of results
• Play and danmaku counts returned both as shown on the site and parsed to integers
• Per keyword error handling: one bad term never fails the whole run
• Scheduling friendly, so your dataset stays current

Input

Field	Type	Required	Default	Description
keywords	array of strings	Yes		Search terms to look up on Bilibili. Results from all keywords land in one dataset.
maxItems	integer	No	33	Maximum number of videos to collect per keyword. Up to 1000.
requestTimeoutSecs	integer	No	30	Per request timeout in seconds.

Example input

{
    "keywords": ["哈利波特", "原神", "英雄联盟"],
    "maxItems": 33
}

With this input, the actor collects up to 33 videos for each of the three keywords, for up to 99 results total.

Output

The actor writes one record per video to the dataset. Example record:

{
    "keyword": "哈利波特",
    "bvid": "BV1eRhg6NEc4",
    "videoUrl": "https://www.bilibili.com/video/BV1eRhg6NEc4/",
    "videoTitle": "哈利波特来B站了",
    "coverUrl": "https://i1.hdslb.com/bfs/archive/8ddcc00f2929aa4b4beece5b089ef14e76d4b887.jpg",
    "author": "飞毛腿_摩根",
    "authorUrl": "https://space.bilibili.com/390186540",
    "authorId": "390186540",
    "plays": "3407",
    "playsCount": 3407,
    "danmaku": "7",
    "danmakuCount": 7,
    "duration": "04:05",
    "publishedText": "5小时前",
    "scrapedAt": "2026-08-25T10:58:39.856865+00:00"
}

Output fields

Field	Type	Description
keyword	string	The search term this result came from.
bvid	string	Bilibili video ID (BV code).
videoUrl	string	Direct link to the video.
videoTitle	string	Video title.
coverUrl	string	Thumbnail image URL.
author	string	Uploader name.
authorUrl	string	Uploader channel URL.
authorId	string	Uploader numeric ID.
plays	string	Play count as shown on the site, for example 43.2万.
playsCount	integer	Play count parsed to a plain number.
danmaku	string	Danmaku (on screen comment) count as shown on the site.
danmakuCount	integer	Danmaku count parsed to a plain number.
duration	string	Video length, for example 04:05.
publishedText	string	Publish date as shown on the site, for example 5小时前 or 07-31.
scrapedAt	string	UTC timestamp of when the result was collected.

If a keyword fails, the actor still writes a record for it with an error field and a scrapedAt timestamp, so nothing is silently dropped.

How to use

1. Open the actor and enter one or more keywords.
2. Set Max items per keyword if you want more or fewer than the default of 33.
3. Click Start.
4. When the run finishes, download the results as JSON, CSV, or Excel, or read them from the dataset API.

Run on a schedule

Use Apify Schedules to run the actor every hour, day, or week. Each run appends fresh results, which is useful for tracking how rankings and view counts change over time.

Get results from the API

Every run stores results in a dataset you can fetch directly:

https://api.apify.com/v2/datasets/{datasetId}/items?format=json

Tips and notes

• Chinese keywords work best for Chinese content, since Bilibili is a China focused platform. English terms also work.
• Play and danmaku counts on Bilibili use 万 (ten thousand) and 亿 (hundred million). The actor keeps the original text and also gives you a parsed integer in playsCount and danmakuCount.
• Publish dates are shown relative (6小时前, meaning 6 hours ago) or as MM-DD for older videos, matching how Bilibili displays them.
• Per keyword limits mean a run with 5 keywords and a limit of 33 can return up to 165 records.
• Higher limits take longer. The first page of each keyword loads quickly. To go beyond roughly the first 35 results, the actor loads deeper result pages in full, which adds a few seconds per page. A limit of 100 typically fetches three pages per keyword.

Frequently asked questions

Do I need a Bilibili account? No. The actor does not require any login.

Can I search many keywords at once? Yes. Add as many as you like to the keywords list. Each is searched independently and the limit applies to each one.

What if a keyword returns no results? The run continues with the next keyword. Empty results simply add no records.

Which formats can I export? JSON, CSV, Excel, HTML, RSS, and more through the dataset export options and API.