Bilibili Video Scraper (哔哩哔哩 / B站)

Bilibili Video Scraper extracts public Bilibili videos – search results, the popular (热门) board and full stats for any video URL – into clean JSON, CSV or Excel. No login, no cookies, $3.50 per 1,000 videos.

What can you use it for?

• Market & trend research in China – see which products, games, brands and topics get views, likes and danmaku on B站.
• Influencer & KOL discovery – find the uploaders (UP主) behind the most-watched videos for a keyword.
• Brand monitoring – track new videos about your brand or competitors, sorted by newest.
• Content strategy – compare titles, durations, tags and engagement of top videos in your niche.
• AI / LLM pipelines – feed structured video data (titles, descriptions, tags, stats) to your agents.

What data do you get?

Field	Description
bvid, aid, url	Video id (BV…), numeric id and link
title, description, tags	Text of the video (HTML highlighting removed)
coverUrl	Cover image (HTTPS)
durationSec	Duration in seconds
publishedAt	Publish time (ISO 8601, UTC)
views, likes, danmaku, comments, favorites, coins, shares	Engagement numbers (coins & shares: popular, video URLs, or search with Add full stats)
category	Bilibili category (e.g. 汽车, 游戏)
authorName, authorId, authorUrl	Public channel of the uploader
source, searchQuery, rank	How the video was found and its position
scrapedAt	When the row was collected

How to use it

1. Choose What to scrape: Search videos by keyword, Popular / trending videos or Video details from URLs.
2. Enter keywords (Chinese or English work) or video links.
3. Set Max videos per keyword and, for search, the sort order (best match, most views, newest, most danmaku, most favorites).
4. Click Start and download the results as JSON, CSV, Excel or HTML – or get them through the API.

Example input

{
  "mode": "search",
  "searchQueries": ["tesla model y"],
  "maxResultsPerQuery": 20,
  "sortBy": "totalrank",
  "includeDetails": true
}

Example output (one item, uploader anonymised)

{
  "bvid": "BV1s2421c7zm",
  "aid": 1600435225,
  "url": "https://www.bilibili.com/video/BV1s2421c7zm",
  "title": "开箱特斯拉 Model Y",
  "description": null,
  "coverUrl": "https://i1.hdslb.com/bfs/archive/7660ce99551c8f7914102702bed64e192ed3133d.jpg",
  "durationSec": 529,
  "publishedAt": "2024-02-04T12:34:25.000Z",
  "views": 111473,
  "danmaku": 25,
  "likes": 297,
  "coins": 40,
  "favorites": 832,
  "shares": 14,
  "comments": 25,
  "category": "模玩·周边",
  "tags": ["玩具", "模玩跨年总动员", "汽车模型", "开箱", "我要做模王！", "模型"],
  "authorName": "Example UP主",
  "authorId": 123456789,
  "authorUrl": "https://space.bilibili.com/123456789",
  "source": "search",
  "searchQuery": "tesla model y",
  "rank": 3,
  "scrapedAt": "2026-10-28T19:39:09.786Z"
}

API (JavaScript)

import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: 'YOUR_APIFY_TOKEN' });
const run = await client.actor('flarewow/bilibili-video-scraper').call({ mode: 'search', searchQueries: ['原神'], maxResultsPerQuery: 50 });
const { items } = await client.dataset(run.defaultDatasetId).listItems();

API (Python)

from apify_client import ApifyClient
client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("flarewow/bilibili-video-scraper").call(run_input={"mode": "popular", "maxResultsPerQuery": 100})
items = client.dataset(run["defaultDatasetId"]).list_items().items

Pricing

$3.50 per 1,000 videos (pay per result, $0.0035 per video) plus a tiny Apify start fee. You pay only for videos saved to the dataset – failed or empty runs cost nothing for results. Add full stats does not change the price.

Limits & FAQ

• How many results per keyword? Bilibili search returns up to ~1,000 videos (50 pages × 20) per keyword; use several keywords or sort orders for more.
• Popular board: up to ~1,000 videos from 热门.
• Comments and danmaku text are not collected (they contain data of private users).
• Deleted, private or region-locked videos are reported in the run's ERRORS record and skipped.
• Blocked requests are retried automatically with a new session. If nothing could be scraped, the run fails with a clear message instead of finishing "successful" with an empty dataset.
• Proxy: Apify Proxy (datacenter) is used by default; video detail pages automatically use residential IPs because Bilibili blocks datacenter IPs there.

Integrations

Works with Make, Zapier, n8n, Google Sheets, Slack, webhooks and the Apify MCP server (use it as a tool in Claude, ChatGPT or your own AI agent). Schedule it to monitor new videos for your keywords daily.

Responsible use

The actor collects only publicly available video information and public channel names/links of uploaders. It does not log in, does not collect comments or private-user data, and does not bypass CAPTCHAs. Use the data in line with Bilibili's terms and the laws that apply to you (e.g. GDPR, PIPL).

---

中文说明

B站视频采集器：按关键词搜索视频、采集热门榜、或根据视频链接获取完整数据（播放、点赞、投币、收藏、分享、弹幕、评论数、标签、UP主公开主页）。无需登录、无需 Cookie，结果可导出为 JSON / CSV / Excel，价格 每 1000 条视频 3.5 美元。

字段	含义
bvid / url	BV号 / 视频链接
title / description / tags	标题 / 简介 / 标签
views / likes / coins / favorites / shares	播放 / 点赞 / 投币 / 收藏 / 分享
danmaku / comments	弹幕数 / 评论数
durationSec / publishedAt	时长（秒）/ 发布时间
category	分区
authorName / authorUrl	UP主名称 / 主页

使用方法：选择模式（关键词搜索 / 热门 / 视频链接）→ 输入关键词或链接 → 设置每个关键词的数量和排序（综合排序、最多播放、最新发布、最多弹幕、最多收藏）→ 运行。仅采集公开数据，不采集评论内容和普通用户的个人信息。