Bilibili Scraper – Videos, Search, Danmaku & Top 100 Ranking
Pricing
from $3.00 / 1,000 videos
Bilibili Scraper – Videos, Search, Danmaku & Top 100 Ranking
Scrape Bilibili (哔哩哔哩 / B站) videos by URL, keyword search or Top 100 ranking: views, likes, coins, favorites, tags, channel, plus danmaku (bullet comments) with reaction peaks. No login, no cookies. $3 per 1,000 videos.
Pricing
from $3.00 / 1,000 videos
Rating
0.0
(0)
Developer
CN Data Lab
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
25 days ago
Last modified
Categories
Features
Share
Bilibili Scraper – Videos, Search, Danmaku (Bullet Comments) & Top 100 Ranking
Bilibili Scraper extracts public data from Bilibili (哔哩哔哩 / B站), China's largest video platform for anime, gaming, tech and Gen Z culture. It needs no login, no cookies and no browser. Give it video URLs, search keywords or the Top 100 ranking, and get structured JSON/CSV/Excel with full engagement stats and, optionally, danmaku (弹幕, timed bullet comments).
What you can scrape
- 🎬 Videos by URL,
BV…/av…ID, mobile link orb23.tvshort link - 🔎 Keyword search (Chinese or English), up to 1,000 results per keyword, sorted by relevance, views, newest, danmaku or favorites
- 🏆 Official Top 100 ranking (排行榜), all categories or one category
- 📊 Stats per video: views, likes, coins, favorites, shares, danmaku count, comment count, tags, duration, publish date, category, channel name/ID/URL
- 💬 Danmaku (optional): every message as its own item with video timestamp, send time, mode and color, plus a per-video reaction timeline and top-5 peak moments
- 🗨️ Top comments preview (optional, free): up to 3 hot comments per video, as shown to logged-out visitors
Use cases
| Use case | How |
|---|---|
| Brand and market research in China | Search your product, game or competitor and read reactions moment by moment |
| Game and media studios | Measure reactions to trailers and launches with danmaku peaks |
| Influencer / KOL discovery | Find top channels for a keyword by views, coins and favorites |
| AI / NLP datasets | Timestamped Chinese short-text (danmaku) |
| Trend monitoring | Schedule the Top 100 ranking and track what goes viral |
| Academic research | Online culture and sentiment studies |
How to use
- Click Try for free (the input is pre-filled with one video and one search).
- Add video URLs, search keywords and/or enable Top 100 ranking. Turn on Scrape danmaku if you want bullet comments.
- Click Start, then download JSON, CSV, Excel or HTML, or read the results through the API.
Input example
{"searchKeywords": ["黑神话 悟空"],"maxVideosPerKeyword": 20,"searchOrder": "click","videoUrls": ["https://www.bilibili.com/video/BV1xx411c7mD"],"includeRanking": false,"scrapeDanmaku": true,"maxDanmakuPerVideo": 500}
| Field | Description |
|---|---|
videoUrls | Video URLs, BV…/av… IDs, mobile links, b23.tv short links |
searchKeywords, maxVideosPerKeyword, searchOrder | Keyword search. Order: totalrank, click, pubdate, dm, stow |
includeRanking, rankingCategoryId, maxRankingVideos | Top 100 ranking. 0 = all categories; e.g. 4 games, 188 tech, 36 knowledge, 3 music, 1 anime, 211 food |
scrapeDanmaku, maxDanmakuPerVideo, danmakuTimelineBucketSec | Danmaku on/off (default off), cap per video (default 1,000), timeline bucket size (default 30 s) |
includeTags | Add uploader tags (default on) |
includeTopComments, includeCommentAuthor | Top-comment preview (default off); commenter nicknames (default off) |
maxConcurrency, proxyConfiguration | Parallel requests (default 3); optional proxy, normally not needed |
Output example
Every dataset item has a type: video, danmaku or error. The dataset has ready-made Videos and Danmaku views. Dates are ISO 8601 in UTC.
Video item (real run, shortened):
{"type": "video","bvid": "BV1A4421f7nX","url": "https://www.bilibili.com/video/BV1A4421f7nX","title": "【完结】《黑神话:悟空》全程无伤全收集全成就全支线全结局全BOSS无伤 | 完美剧情流程攻略-4K…","coverUrl": "https://i0.hdslb.com/bfs/archive/7560426b7ad02534806b8fb695f437d1cc4d6c0b.png","durationSec": 100357,"publishedAt": "2024-08-20T17:40:22+00:00","partsCount": 59,"channelId": 288742222,"channelName": "劍訣浮雲","channelUrl": "https://space.bilibili.com/288742222","views": 1394920,"danmakuCount": 20522,"comments": 2086,"likes": 29524,"coins": 23907,"favorites": 24402,"shares": 3242,"tags": ["西游记", "孙悟空", "黑神话悟空", "黑神话", "游戏科学", "国产3A"],"danmakuScraped": 300,"danmakuPeaks": [{"startSec": 210, "count": 21}, {"startSec": 630, "count": 16}],"danmakuTimeline": [{"startSec": 0, "count": 15}, {"startSec": 30, "count": 14}],"source": "search","searchKeyword": "黑神话 悟空","searchRank": 2,"scrapedAt": "2026-10-18T07:56:57+00:00"}
Danmaku item (real run):
{"type": "danmaku","bvid": "BV1xx411c7mD","cid": 62131,"danmakuId": "12338673447075844","text": "唯一活着的老片","videoTimeSec": 0.0,"postedAt": "2019-02-20T15:25:26+00:00","mode": 5,"fontSize": 25,"color": "#ffd302"}
mode: 1–3 scrolling, 4 bottom, 5 top, 7 advanced.
Pricing (pay per event)
| Event | Price |
|---|---|
| Video (one video item with stats) | $0.003 ($3 per 1,000) |
| Danmaku (one bullet comment) | $0.0002 ($0.20 per 1,000) |
| Actor start | $0.00005 |
Failed or deleted videos come back as error items and are not charged. Top comments and tags are free. The Actor respects your maximum cost per run and stops cleanly when it is reached.
Examples: 1,000 videos from search ≈ $3. 5 videos with 1,000 danmaku each ≈ $0.015 + $1.
FAQ
Do I need a Bilibili account or cookies? No. The Actor only reads data that Bilibili shows to logged-out visitors.
How fast is it? In our tests, about 380 videos with stats took roughly 2.5 minutes, and 22 videos plus ~13,000 danmaku took under 40 seconds, at the default concurrency. Speed depends on Bilibili's response times.
How many danmaku can I get per video? Danmaku are read in 6-minute segments across the video timeline, up to your maxDanmakuPerVideo cap. You get what Bilibili serves publicly for each segment, which can be lower than the video's total danmakuCount.
Can I scrape comments? Only a preview. Bilibili shows logged-out visitors at most a few hot comments, so the Actor returns up to 3 per video. Full comment threads are not supported. Danmaku are the better source of audience reactions.
Can it scrape user/uploader profiles? No. You get the channel name, ID and URL of each video, but not follower counts, an uploader's video list or other profile data.
Is it legal? The Actor collects only publicly available data, does not log in and does not store commenter identities by default. You are responsible for using the data in line with Bilibili's terms and applicable laws (e.g. GDPR, China's PIPL). If unsure, ask a lawyer.
Can I schedule it or call it from other tools? Yes. Use Apify schedules, webhooks, the API, integrations such as Make, Zapier, n8n and Google Sheets, or an MCP client.
Limitations
- Public, logged-out data only. No private or login-only content, no video files or subtitles.
- Danmaku are collected for the first part (P1) of multi-part videos. Video stats cover the whole video.
- Search returns what Bilibili's search API returns (up to 1,000 results per keyword); the ranking holds at most 100 videos.
- Bilibili may rate-limit or block some IPs. The Actor retries and refreshes its session; if you still see errors, enable Apify Proxy in the input.
- This Actor uses unofficial web endpoints that Bilibili can change without notice. It is not affiliated with Bilibili.
Related Actors
Weibo Scraper for Weibo Hot Search, posts and comments. Google Trends Scraper for keywords, related queries and regional interest.
Feedback
Found a bug or need a feature? Open an issue on the Actor's Issues tab.