Bilibili Creator Leaderboard Scraper

Scrape Bilibili trending video leaderboards across 32 content categories. Get rank position, full video stats (views, likes, coins, danmaku/bullet comments), creator details, and a crawl timestamp — ready for time-series trend analysis.

What does Bilibili Creator Leaderboard Scraper do?

This actor hits the Bilibili public ranking API (api.bilibili.com/x/web-interface/ranking/v2) for each requested category and extracts the top 100 videos. Each record includes:

• Rank position within the category (1–100)
• Video metadata: BV ID, title, description, duration, publish timestamp, cover image, first-frame thumbnail, short link
• Engagement stats: view count, like count, coin count, favorite count, danmaku (bullet comment) count, reply count, share count
• Creator info: user ID (mid), display name, avatar URL
• Location signal: IP-province annotation
• Snapshot timestamp: ISO 8601 crawl time for time-series stitching

Use cases

• Trend monitoring: Track which videos are rising and falling across categories daily or hourly
• MCN analytics: Identify breakout creators before they reach peak visibility
• Brand monitoring: Spot trending content in relevant verticals (Knowledge, Digital, Gaming, etc.)
• Market research: Benchmark content performance metrics across Bilibili's major categories
• Training data: Build ranked video datasets for recommendation model training

Input

Parameter	Type	Required	Description
maxItems	integer	Yes	Maximum records to collect across all categories. Default: 10
categories	array	Yes	List of category TIDs to scrape. Default: [0] (All).

Category TID reference

TID	Category
0	All (全部)
1	Animation (动画)
3	Music (音乐)
4	Gaming (游戏)
17	Entertainment (娱乐)
36	Knowledge (知识)
95	Digital (数码)
119	Lifestyle (生活)
129	Fashion (时尚)
167	Car (汽车)
174	Sports (运动)
176	Film & TV (影视)
181	TV Series (剧集)
188	Documentary (纪录片)

Output

Each record in the dataset:

{
  "rank_position": 1,
  "category_id": 0,
  "category_name": "全部",
  "bvid": "BV1JiVb6EEi7",
  "aid": 116655892010800,
  "title": "【毕导】这是什么妖术？酒在坛子里是绿色，舀出来瞬间变红！",
  "description": "一个诡异的卖酒视频...",
  "duration_seconds": 760,
  "publish_timestamp": 1780027992,
  "creator_mid": 254463269,
  "creator_name": "毕导",
  "creator_avatar_url": "https://i2.hdslb.com/bfs/face/...",
  "cover_url": "http://i1.hdslb.com/bfs/archive/...",
  "score": 0,
  "view_count": 7044922,
  "danmaku_count": 20659,
  "reply_count": 7075,
  "favorite_count": 54402,
  "coin_count": 93064,
  "share_count": 10561,
  "like_count": 354797,
  "pub_location": "浙江",
  "short_link": "https://b23.tv/BV1JiVb6EEi7",
  "first_frame_url": "http://i1.hdslb.com/bfs/storyff/...",
  "tag_v2_name": "科学科普",
  "snapshotted_at": "2026-05-31T10:58:45.503Z"
}

Technical notes

• The Bilibili ranking API returns exactly 100 items per category. Each category is a separate API call.
• Requests use browser-like Referer and Origin headers to avoid Bilibili's risk-control response (code: -352).
• A 1.1-second delay between requests prevents rate limiting.
• No proxy required — the public API works from datacenter IPs with the correct headers.
• Rankings are time-sensitive snapshots: the snapshotted_at field enables time-series stitching across runs.