Bilibili Scraper 🎬 — Videos, Comments, Danmaku & Creators (B站 / 哔哩哔哩)

Scrape Bilibili video data, engagement metrics, comments, danmaku (弹幕) and creator / UP主 profiles at scale — by keyword, URL or trending feed. No API key. No login. No account.

This Bilibili scraper turns the public data on Bilibili (哔哩哔哩, a.k.a. B站 or bili bili) — one of China's largest video platforms, home of danmaku bullet comments and a massive creator (UP主) economy — into clean, structured JSON / CSV / Excel you can actually use. It's a practical, drop-in Bilibili API alternative that just works: paste a keyword, a video link, or a creator page, press Start, and get rows.

Export to JSON, CSV, Excel, or API — straight into Sheets, a database or your AI pipeline.

Built for market researchers, KOL / influencer agencies, brand & social-listening teams, AI/NLP data engineers, MCNs and analysts who need Bilibili data without the hassle.

---

📑 Table of contents

• Why scrape Bilibili
• What this Bilibili scraper does
• Use cases
• Input
• Output
• Pagination & scale
• How to use
• Tips
• FAQ
• Related scrapers
• Legal & responsible use

---

🔎 Why scrape Bilibili?

Bilibili data is one of the richest, least-tapped signals on the Chinese internet. With 300M+ monthly active users, B站 is where Chinese audiences discuss games, tech, anime, food, study and daily life — and where the famous danmaku (弹幕) bullet comments scroll across every video, giving you raw, real-time audience reaction you can't get anywhere else.

Collecting it by hand is painful. Bilibili has no open public API for bulk export, the site is in Chinese, result pages are WBI-signed and guarded by risk-control, and the engagement stats you actually want — views, danmaku, likes, coins (投币), favorites (收藏), comments — are scattered across different endpoints. Copy-pasting video by video does not scale.

This Bilibili scraper does it for you: it searches, paginates, de-duplicates and returns tidy flat records with every metric already parsed — plus optional comments, danmaku and full creator profiles with business emails for lead-gen.

Common reasons people scrape Bilibili:

• 📈 Track trends on B站 — which topics, games, creators and formats are winning right now.
• 🤝 Build UP主 lead lists — rank creators by real followers and pull the business email from their bio.
• 🏷️ Monitor brand sentiment — read the danmaku and comments on your product videos and competitors'.
• 🤖 Harvest Chinese-language AI data — authentic comments + danmaku for NLP, sentiment and LLM training.
• 🎬 Benchmark content — compare coins, favorites and danmaku, not just view counts.
• 🎓 Research danmaku culture, fandoms and online behavior at academic scale.
• 📊 Feed dashboards and notebooks with fresh Bilibili data on a schedule.

---

✨ What this Bilibili scraper does

• ✅ Four ways to find videos — 🔎 keyword search (English or 中文), 🎬 direct video URL / ID (BV / av), 👤 by creator (UP主), and 🔥 the trending / popular feed. Mix them freely in one run.
• ✅ Automatic pagination — walks search and trending pages for you and returns hundreds to thousands of de-duplicated records (by bvid) in a single run. A broad two-keyword run returns ~500 videos in under a minute.
• ✅ Full engagement metrics — views (播放), danmakuCount (弹幕), likes (点赞), favorites (收藏), comments (评论), plus coins (投币) and shares on the trending feed.
• ✅ Rich video fields — title, description, cover thumbnail, category (分区), durationSeconds, publishedAt, tags, canonical url, bvid / aid, and the author's name, mid and space URL.
• ✅ Creator / UP主 profiles — followers, following, videoCount, totalLikes, bio, level, verification / official title, avatar, and a business email auto-detected from the bio for outreach.
• ✅ Comments (optional) — top comments sorted by likes, with author, like count, reply count and timestamp.
• ✅ Danmaku / 弹幕 (optional) — Bilibili's signature on-screen bullet comments with their text, position in the video (seconds) and when they were posted. Unique to Bilibili.
• ✅ Clean, decoded text — HTML entities (", &, ' …) are decoded so titles, comments and bios read naturally.
• ✅ Fast & pay-as-you-go — pure HTTP, no heavy browser, parallel fetching, automatic proxy rotation and retry on Bilibili's risk-control. You only pay for the rows you keep.

💡 The danmaku difference. Most Bilibili tools stop at views and titles. This one gives you the bullet comments that scroll across every video — the richest, rawest audience-reaction signal on the Chinese internet. Perfect for sentiment, trends and AI training data.

---

🎯 Use cases

• Chinese market research — which topics, formats and creators win on B站, with real engagement numbers.
• KOL / influencer discovery — rank UP主 by followers, then outreach via the business email in their bio.
• Brand & social listening — danmaku + comments on your product videos and competitors' for sentiment.
• AI / ML & NLP datasets — large, authentic Chinese comment and danmaku corpora for sentiment and LLM work.
• Creators & MCNs — benchmark competitor videos by coins, favorites and danmaku; find content gaps.
• Analysts & data teams — feed dashboards, notebooks and databases with structured B站 data as CSV/JSON.
• Academic & social studies — study danmaku culture, fandoms and online behavior at scale.

---

📥 Input

Fill in any mix of the fields below — or just press Start to grab a sample of trending videos. All inputs are optional.

Field	Type	Description
searchKeywords	array of strings	Find videos by keyword — English or Chinese (e.g. python, 美食, 原神). One per line.
videoUrls	array of strings	Scrape specific videos. Full URLs (https://www.bilibili.com/video/BV1...) or bare IDs (BV1... / av170001).
userIds	array of strings	Creator (UP主) profiles. Space URLs (https://space.bilibili.com/123456) or bare numeric IDs.
scrapePopular	boolean	Also pull the current Bilibili trending / popular feed. Default false.
popularCount	integer	How many trending videos to fetch (1–200). Default 30.
includeComments	boolean	Attach top comments (sorted by likes) to each video. Default false.
maxCommentsPerVideo	integer	Upper limit of comments per video (1–1000). Default 20.
includeDanmaku	boolean	Attach danmaku (bullet comments) with text + timestamp. Default false.
maxDanmakuPerVideo	integer	Upper limit of danmaku per video (1–6000). Default 200.
searchPages	integer	Result pages to pull per keyword (~42 videos/page, 1–50). Default 2.
searchOrder	string	Sort search results: totalrank (best match), click (most viewed), pubdate (newest), dm (most danmaku), stow (most favorited).
maxVideos	integer	Overall cap on videos across all sources (1–10,000). Default 200.
maxConcurrency	integer	Videos enriched in parallel (1–20, recommended 6–12). Default 8.
proxyConfiguration	object	Proxy settings. The default Apify Proxy works out of the box for every mode.

Example — search two topics and keep the top 100 videos with comments:

{
  "searchKeywords": ["机器学习", "python tutorial"],
  "searchPages": 2,
  "searchOrder": "totalrank",
  "includeComments": true,
  "maxCommentsPerVideo": 20,
  "maxVideos": 100,
  "proxyConfiguration": { "useApifyProxy": true }
}

Deep-dive specific videos, danmaku included:

{
  "videoUrls": ["https://www.bilibili.com/video/BV1GJ411x7h7", "BV1uMTY6aEZv"],
  "includeDanmaku": true,
  "maxDanmakuPerVideo": 1000
}

Build a creator (UP主) lead list:

{ "userIds": ["https://space.bilibili.com/431073645", "2267573"] }

Grab what's trending right now (full stats including coins & shares):

{ "scrapePopular": true, "popularCount": 100 }

---

📤 Output

Each result is one flat record. Video records carry full engagement metrics; creator records carry profile + lead-gen fields. Below is a real video record from a keyword-search run:

{
  "recordType": "video",
  "bvid": "BV1rpWjevEip",
  "aid": 113006243481679,
  "cid": null,
  "url": "https://www.bilibili.com/video/BV1rpWjevEip",
  "title": "【全748集】目前B站最全最细的Python零基础全套教程，2026最新版",
  "description": "本套教程从零开始讲解，手把手教学，包含基础语法、进阶语法、爬虫、自动化办公、数据分析…",
  "cover": "https://i2.hdslb.com/bfs/archive/a979056b1a32012cdd00d48fbc3732d253e30620.jpg",
  "category": "计算机技术",
  "durationSeconds": 143894,
  "publishedAt": "2024-08-22T14:59:18.000Z",
  "views": 21406006,
  "danmakuCount": 135253,
  "likes": 590100,
  "coins": null,
  "favorites": 907462,
  "shares": null,
  "comments": 367891,
  "author": "Python官方课程",
  "authorMid": 3546597933714079,
  "authorUrl": "https://space.bilibili.com/3546597933714079",
  "authorAvatar": "https://i1.hdslb.com/bfs/face/6822e4fdc8f82c69e10accd1ab11a109702bccbf.jpg",
  "tags": "程序员,Python教程,Python入门零基础,计算机,编程语言,数据分析",
  "sourceType": "search",
  "sourceQuery": "python",
  "scrapedAt": "2026-11-02T00:35:19.056Z"
}

A creator (UP主) record:

{
  "recordType": "creator",
  "mid": "431073645",
  "name": "你的影月月",
  "spaceUrl": "https://space.bilibili.com/431073645",
  "followers": 10138744,
  "following": 312,
  "videoCount": 2190,
  "totalLikes": 88450231,
  "bio": "攻略在合集观看！商务合作邮箱 jyybiz@qq.com",
  "businessEmail": "jyybiz@qq.com",
  "level": 6,
  "official": "2025百大UP主、知名游戏UP主",
  "avatar": "https://i2.hdslb.com/bfs/face/....jpg",
  "scrapedAt": "2026-07-26T10:00:00.000Z"
}

🎬 Video field reference

Field	Description
recordType	"video" for video records
bvid / aid	Bilibili video identifiers
cid	Internal play ID — present on trending-feed videos; null for search results (used internally for danmaku)
url	Canonical video link
title	Video title (HTML entities decoded)
description	Video description — present when the uploader provides one (~77% of search results)
cover	Thumbnail image URL (https)
category	Bilibili category (分区), e.g. 计算机技术, 单机游戏, 数码
durationSeconds	Length in seconds
publishedAt	Publish date (ISO 8601)
views	Play count (播放)
danmakuCount	Number of danmaku (弹幕)
likes	Likes (点赞)
favorites	Favorites (收藏)
comments	Comment / reply count (评论)
coins	Coins given (投币) — trending feed only; null for keyword search
shares	Shares (分享) — trending feed only; null for keyword search
author / authorMid / authorUrl / authorAvatar	Creator name, numeric ID, space link and avatar
tags	Comma-separated video tags (search results)
sourceType / sourceQuery	How the record was found (search / popular / video) and the originating keyword/ID
topComments[]	Top comments — only when Include comments is on
danmaku[]	Bullet comments — only when Include danmaku is on
scrapedAt	ISO 8601 collection timestamp

ℹ️ Why some metrics differ by mode. Bilibili's keyword-search endpoint returns views, danmaku, likes, favorites and comments for every result, but not coins, shares or cid. Its trending/popular feed returns the complete stat object (all of the above at 100%). Run scrapePopular when you specifically need coins and shares.

👤 Creator (UP主) field reference

Field	Description
recordType	"creator" for creator records
mid / spaceUrl	Creator ID and profile link
name	Display name
followers / following	Follower count (粉丝) and accounts followed
videoCount	Number of uploads
totalLikes	Lifetime likes received
bio	Profile signature (签名) — present when set
businessEmail	Business email detected in the bio — present when the creator lists one; great for outreach
level	Bilibili level (0–6)
official	Verification / official title — present for verified creators
avatar	Avatar image URL

💬 Comment & 🌊 danmaku fields

Field	Belongs to	Description
text	both	Comment / danmaku text (entities decoded)
author, authorMid, likes, replyCount, time	comment	Author, engagement and timestamp
videoTimeSec, sentAt	danmaku	Position in the video (seconds) and when it was posted

---

🔁 Pagination & scale

This Bilibili scraper paginates automatically so you can pull hundreds to thousands of de-duplicated records in a single run — no manual page-clicking.

• 🔎 Keyword search walks result pages for every keyword (~42 videos per page). Set Search pages per keyword (searchPages, 1–50) to go deeper. Requests are WBI-signed and automatically retried on a fresh IP if Bilibili's risk-control challenges a request.
• 🔥 Trending feed pages through the popular endpoint (20 per page) up to Number of trending videos (popularCount).
• 💬 Comments and 🌊 danmaku paginate per video up to Max comments per video / Max danmaku per video, so you decide how deep each enrichment goes.
• Every video is de-duplicated by bvid across all sources, then capped by Max videos (maxVideos, up to 10,000). Creators, keywords, URLs and the trending feed can all run in the same job.

Because the actor is pure HTTP (no browser), pages are fetched fast and in parallel (Max concurrent requests, default 8), and each request rotates through the Apify Proxy so large runs stay reliable. A verified benchmark run of 2 keywords × 7 pages returned 498 unique videos in ~58 seconds with 0 duplicates and 100% core-field coverage. You only pay for the rows you keep.

---

⚙️ How to use

1. Open the actor on the Apify Console and go to the Input tab.
2. Add any mix of 🔎 Search keywords, 🎬 Video URLs, 👤 Creators, or tick 🔥 Include trending videos.
3. (Optional) Turn on 💬 Include comments and/or 🌊 Include danmaku and set their per-video limits.
4. Set Max videos and Search pages per keyword to control run size.
5. Press Start ▶️ and watch records stream into the dataset.
6. Export as JSON, CSV, Excel, HTML or RSS, or pull via the Apify API.
7. Automate it: add a Schedule to refresh a watchlist, or trigger via API / webhooks and pipe into Make, Zapier, Google Sheets or your database.

---

💡 Tips

• Need coins & shares? Add scrapePopular: true — the trending feed carries the complete stat object (search does not).
• Search in Chinese for the richest results — queries like 原神, 美食, 编程 surface far more native content than their English equivalents.
• Go deep on a topic by raising searchPages (each page ≈ 42 videos); combine several keywords in one run and let maxVideos cap the total.
• Change the ranking with searchOrder — pubdate for the freshest uploads, click for the most-viewed, dm for the most danmaku.
• Keep runs fast and cheap by leaving comments/danmaku off until you need them — they add per-video requests.
• Lead-gen shortcut: run creator mode on a list of userIds and filter the output where businessEmail is present.

---

❓ FAQ

---

🔗 Related scrapers

Building a cross-platform social, creator or video dataset? Pair this Bilibili scraper with:

• YouTube Contact Scraper — pull creator emails and channel contacts from YouTube.
• YouTube Comments Scraper — export comment threads and engagement from YouTube videos.
• Xiaohongshu (RedNote) Scraper — notes, creators and engagement from China's RED / 小红书.
• WeChat Official Account Scraper — articles and metrics from WeChat 公众号 publishers.
• Patreon Scraper — creators, tiers and membership data from Patreon.
• Linktree Scraper — expand creator bio links into structured contact and social profiles.

---

⚖️ Legal & responsible use

This tool collects publicly available information only — the same data any visitor can see without logging in. It does not access private accounts, bypass paywalls, or collect anything behind authentication.

You are responsible for how you use the data. Please:

• Respect Bilibili's Terms of Service and robots policies.
• Comply with applicable laws, including privacy and data-protection rules (e.g. GDPR, CCPA, PIPL) when handling any personal data such as creator contact details.
• Use collected data lawfully — for research, analytics and legitimate business purposes — and not for spam, harassment or re-identification.

If you're unsure whether your use case is permitted, seek legal advice. This actor is provided for lawful data-collection purposes only.

---

🌏 Also searched as

Bilibili scraper · B站 scraper · 哔哩哔哩 爬虫 · bilibili video scraper · bilibili comments scraper · danmaku scraper · 弹幕 scraper · bilibili UP主 scraper · bilibili creator / influencer scraper · bilibili API · bilibili data export · bili bili scraper · Chinese video data

---

Made for researchers, marketers, agencies and data teams who need Bilibili data without the hassle. Questions or a field you'd like added? Open an issue on the actor's Issues tab.