Bilibili (哔哩哔哩) Scraper

BILIBILI (哔哩哔哩) SCRAPER  •  BILIBILI API See what the audience said, second by second. Search Bilibili by keyword or paste a video link and get videos with plays, likes, coins, favourites, danmaku and comment counts, the UP主 behind each one, and the full comment threads. Plus the two data sources that are Bilibili's own: raw danmaku (弹幕) bullets timestamped to the second, and the platform's native subtitle track as a timed transcript. No Bilibili account, no login, no cookies. FAILED LOOKUPS ARE FREE  DANMAKU (弹幕) + NATIVE SUBTITLES

Copy to your AI assistant

Paste this into ChatGPT, Claude, Cursor, or any LLM to start using this Actor right away.

atomus/bilibili-scraper is an Apify Actor that returns Bilibili (哔哩哔哩, B站) data as structured JSON in eight modes, picked with searchType: keyword search, video detail, comments, danmaku (弹幕 bullet-comments), native subtitle transcript, UP主 profile, UP主 videos, and trending videos or hot-search terms. Use it whenever someone needs Bilibili data without cookies, a login or a browser - trend research on Chinese video, audience-reaction analysis second by second from danmaku, transcript mining for summarisation or translation, UP主 (creator) shortlists, or comment analysis. Run it with curl: curl -X POST "https://api.apify.com/v2/acts/atomus~bilibili-scraper/run-sync-get-dataset-items?token=APIFY_TOKEN" -H "Content-Type: application/json" -d '{"searchType":"search","keywords":["编程"],"maxItems":50,"sortOrder":"click"}'. Or in Python: ApifyClient("APIFY_TOKEN").actor("atomus/bilibili-scraper").call(run_input={"searchType":"danmaku","videoUrls":["BV1rpWjevEip"]}) then client.dataset(run["defaultDatasetId"]).list_items().items. Inputs: searchType (search, video-detail, comments, danmaku, transcript, profile, user-videos, trend), keywords (string[], search only), videoUrls (string[], full bilibili.com/video/BV... URLs or bare BV ids, used by video-detail, comments, danmaku and transcript), userUrls (string[], space.bilibili.com URLs or bare numeric UIDs, used by profile and user-videos), maxItems (int, default 20, max 5000, per keyword or target), sortOrder (totalrank, click, pubdate, dm, stow), durationFilter (any, short, medium, long, verylong), publishedAfter and publishedBefore (YYYY-MM-DD), trendType (videos or keywords), includeReplies (bool, fetches the full reply thread under each comment and bills every reply as a comment). Every row carries searchType plus the fields of its mode: video rows have bvid, aid, title, desc, url, cover, duration, category, tags, view_count, like_count, coin_count, favorite_count, share_count, danmaku_count, reply_count, pubdate and a nested owner object; comment rows have rpid, content, like_count, reply_count, ctime, location, up_liked, up_replied, user, user_level, user_vip, user_verified and replies; danmaku rows have text, progress (seconds into the video), mode, color and ctime; transcript rows have available, lang, ai, tracks, segments and text; profile rows have mid, name, sign, level, fans, following, place, verified, verify_desc and vip. Billing is per result and per mode: 0.01 USD per video from keyword search or UP主 videos, 0.02 per video detail, 0.008 per comment with a minimum of 2 per video, 0.012 per UP主 profile, 0.01 per video for danmaku, 0.01 per video for a transcript, and 0.008 per trending result with a minimum of 2; Apify Gold plans and above pay 20 percent less, and failed lookups, videos with no subtitle track and every search filter cost nothing. Full input schema, every enum and default, and the complete output field list: GET https://api.apify.com/v2/acts/atomus~bilibili-scraper/build/default

---

Bilibili MCP Server: use these Actors from ChatGPT, Claude or Cursor

Point your AI assistant at Atomus and it can read the social platforms on its own: Bilibili, the other three big Chinese networks, TikTok and X. No glue code, no scraping logic in your prompts.

{
  "mcpServers": {
    "atomus": {
      "url": "https://mcp.apify.com?tools=atomus/bilibili-scraper,atomus/xiaohongshu-scraper,atomus/douyin-scraper,atomus/weibo-scraper,atomus/tiktok-scraper,atomus/tiktok-comments-scraper,atomus/twitter-scraper",
      "headers": { "Authorization": "Bearer YOUR_APIFY_TOKEN" }
    }
  }
}

Then ask, in plain language:

"Find the most-played Bilibili videos about mechanical keyboards from the last three months, then pull the danmaku on the top one and tell me which moments got the strongest reaction."

That one sentence uses the Actor twice in a row. Pinning the tools= list is what keeps your assistant on these Actors instead of reaching for whatever scraper it finds first.

---

How to scrape Bilibili

Search a keyword, most-played first

{
  "searchType": "search",
  "keywords": ["编程", "机械键盘"],
  "sortOrder": "click",
  "maxItems": 50
}

One row per video, with plays, likes, coins, favourites, danmaku and comment counts. sortOrder accepts totalrank (comprehensive), click (most played), pubdate (newest), dm (most danmaku) and stow (most favourited).

Narrow a search by date and length, for free

{
  "searchType": "search",
  "keywords": ["AI"],
  "publishedAfter": "2026-01-01",
  "publishedBefore": "2026-06-30",
  "durationFilter": "medium",
  "maxItems": 100
}

publishedAfter, publishedBefore and durationFilter are free and are sent to Bilibili as part of the query, so a narrow search genuinely returns and bills fewer videos than a broad one.

Audience reaction, second by second (danmaku / 弹幕)

{
  "searchType": "danmaku",
  "videoUrls": ["https://www.bilibili.com/video/BV1rpWjevEip"]
}

One row per bullet-comment, each with progress — the exact second it appears on screen — so you can line reactions up against the moments that caused them. Billed once per video, not per bullet.

The video's own subtitle track, as a transcript

{
  "searchType": "transcript",
  "videoUrls": ["BV1rpWjevEip"]
}

Returns Bilibili's native subtitle track (including the auto-generated ai-zh one) as timed segments plus the joined full text. There is no audio download and no speech-to-text step, so length does not matter. If the video has no subtitle track you get a row with available: false and nothing is charged.

Every comment on a video, with the full reply threads

{
  "searchType": "comments",
  "videoUrls": ["BV1rpWjevEip"],
  "maxItems": 200,
  "includeReplies": true
}

Each comment carries its IP location (归属地), the commenter's level, VIP and verification, and whether the UP主 liked or replied to it. With includeReplies off you still get up to 3 reply previews per comment for free; with it on, every reply in the thread is fetched and billed as one comment.

A creator (UP主) and their whole catalogue

{
  "searchType": "user-videos",
  "userUrls": ["https://space.bilibili.com/178360345", "178360345"],
  "maxItems": 500
}

Switch searchType to profile on the same input to get the creator card instead: fans, following, level, verification and VIP.

What is trending right now

{ "searchType": "trend", "trendType": "videos", "maxItems": 50 }

trendType: "videos" returns the 热门 feed with each video's recommendation reason; "keywords" returns the 热搜 terms with their rank and heat score.

---

Input

Pick a mode in searchType; every other field belongs to one or more modes and is ignored by the rest.

Parameter	Type	Default	Description
searchType	string	search	The mode: search, video-detail, comments, danmaku, transcript, profile, user-videos, trend.
keywords	string[]	(none)	Search terms. Used by search.
videoUrls	string[]	(none)	Video URLs or bare BV ids. Used by video-detail, comments, danmaku, transcript.
userUrls	string[]	(none)	UP主 space URLs or bare numeric UIDs. Used by profile and user-videos.
maxItems	integer	20	Results per keyword or target, up to 5000. Applies to search, comments, user-videos and trend.
sortOrder	string	totalrank	totalrank, click, pubdate, dm, stow. Free, search only.
durationFilter	string	any	short (<10 min), medium (10–30), long (30–60), verylong (>60). Free, search only.
publishedAfter	string	(none)	YYYY-MM-DD. Free, search only.
publishedBefore	string	(none)	YYYY-MM-DD. Free, search only.
trendType	string	videos	videos (热门) or keywords (热搜). trend only.
includeReplies	boolean	false	💲 comments only. Fetches every reply in a thread; each reply is billed as one comment and counts against maxItems.

⚠️ Which video and creator references work. A video is a full URL (https://www.bilibili.com/video/BV1rpWjevEip) or a bare BV id (BV1rpWjevEip). A creator is a space URL (https://space.bilibili.com/178360345) or a bare numeric UID (178360345). Anything else is skipped with a warning instead of being charged.

maxItems is the main cost lever for the list modes, because they bill per result: a run costs roughly mode price × results × targets. video-detail and profile return one record per target; danmaku and transcript return the whole set from a single call and are billed per video. Duplicate keywords, URLs or UIDs in the same input are collapsed before anything runs, so pasting the same target twice never charges twice.

---

What data does the Bilibili Scraper return?

One row per result. Every row carries searchType so mixed datasets stay sortable; list rows also carry what produced them (keyword, video, uid, trendType).

Group	Fields
Video (search, detail, UP主 videos, trending)	bvid · aid · title · desc · url · cover · duration (s) · category (分区) · tags[] · pubdate · part_count
Video stats	view_count · like_count · coin_count (投币) · favorite_count (收藏) · share_count · danmaku_count · reply_count
Video, detail only	cid · parts[] (每个分P: cid, page, title, duration) · dimension{width,height,rotate} · copyright (1 original, 2 reprint) · dynamic (the UP主's post text)
Video, trending only	rcmd_reason (why it is trending) · pub_location (发布地)
Owner (on every video)	owner.mid · owner.name · owner.avatar
Comment	rpid · content · like_count · reply_count · ctime · root · oid · location (归属地) · up_liked · up_replied · user_level · user_vip · user_verified · user{mid,name,avatar} · replies[] (same shape, nested)
Danmaku (弹幕)	dmid · text · progress (seconds into the video) · mode (1–3 scroll, 4 bottom, 5 top) · fontsize · color (decimal RGB) · ctime · sender_hash
Transcript	available · lang (e.g. ai-zh) · ai · tracks[]{lan,lan_doc,ai,url} · segments[]{from,to,content} · text (joined)
UP主 profile	mid · name · avatar · sign · level · fans · following · friends · place · description · verified · verify_desc · vip · url
Hot search term	rank · keyword · show_name · heat_score · icon

cid is what identifies a video's stream, and it is only present on video-detail rows — the danmaku and transcript modes resolve it for you, at no extra charge.

Example rows

{
  "searchType": "search",
  "keyword": "编程",
  "bvid": "BV1rpWjevEip",
  "aid": "113006243481679",
  "title": "【全748集】Python零基础全套教程",
  "url": "https://www.bilibili.com/video/BV1rpWjevEip",
  "duration": 143894,
  "category": "计算机技术",
  "tags": ["Python教程", "编程语言"],
  "view_count": 18738611, "like_count": 497530, "coin_count": 361759,
  "favorite_count": 774826, "danmaku_count": 123656, "reply_count": 329821,
  "pubdate": 1723442400,
  "owner": { "mid": "3546597933714079", "name": "Python官方课程", "avatar": "https://i2.hdslb.com/..." }
}

{
  "searchType": "danmaku",
  "video": "BV1rpWjevEip",
  "text": "老大 我们真的能学会吗",
  "progress": 71.774, "mode": 1, "color": 16777215, "ctime": 1783262069
}

{
  "searchType": "transcript",
  "video": "BV1rpWjevEip",
  "available": true, "lang": "ai-zh", "ai": true,
  "segments": [{ "from": 0.5, "to": 3.2, "content": "大家好" }],
  "text": "大家好 …"
}

---

How much does it cost to scrape Bilibili?

You pay per result, and each mode is its own event. Apify Gold plans and above pay 20% less on every event.

Mode	Event	Free–Silver	Gold+
Keyword search	video-scraped	$0.01 / video	$0.008
UP主 videos	video-scraped	$0.01 / video	$0.008
Video detail	video-detail-scraped	$0.02 / video	$0.016
Comments (and each reply)	comment-scraped	$0.008 / comment	$0.0064
UP主 profile	profile-scraped	$0.012 / creator	$0.0096
Danmaku	danmaku-scraped	$0.01 / video	$0.008
Transcript	transcript-scraped	$0.01 / video	$0.008
Trending / hot search	trend-scraped	$0.008 / result	$0.0064

Minimum per request: comments bill at least 2 per video, and trending at least 2 per run. One request is one lookup, and that lookup does the same work whether it comes back with 50 rows, 1, or none — the minimum covers the request itself, which is what keeps thin targets worth running instead of being refused.

Free: every search filter (sort, length, date range), the up-to-3 reply previews on each comment, the cid lookup that danmaku and transcript need, failed lookups, and a video that turns out to have no subtitle track — you get the available: false row and are not charged for it.

Free plan: 10 search results per calendar month plus one of every other mode, so you can inspect each row shape before paying. Keyword search and UP主 videos share that same allowance. The counters reset on the 1st, and any paid Apify plan removes the cap.

---

What do people use the Bilibili Scraper for?

• Trend and topic research: search a keyword sorted by plays or danmaku and watch what is rising in a category on 哔哩哔哩.
• Audience reaction analysis: pull raw danmaku with per-second timestamps and find the exact moments a video landed.
• Transcript and content mining: harvest native subtitles for summarisation, translation, or building a searchable index of a channel.
• Creator (UP主) research: fans, level and verification for influencer shortlists, plus the creator's full catalogue with real play counts.
• Comment analysis: themes, sentiment and complaints from a video's comment threads, with the commenter's IP location attached.
• Competitive tracking on the China market: follow a brand's UP主 account and measure what actually gets played.
• Localisation and market research: read how a product, game or anime is discussed by a Chinese audience without an account.
• AI agent context: hand an agent a video's transcript and its danmaku together, so it can reason over what was said and how people reacted.

---

Bilibili Scraper vs cookie-based tools vs the official Bilibili API

	This Bilibili Scraper	Cookie-based scrapers	Official Bilibili API
Bilibili account / cookies	Not needed	Your session cookie required	Chinese phone verification
Breaks when a cookie expires	Never (no cookie used)	Yes	n/a
Access from outside China	Yes	Varies	Restricted
Raw danmaku (弹幕) bullets	Yes, one row each, timestamped	Rare	Manual, WBI-signed
Native subtitle transcript	Yes, timed segments plus full text	Rare	Manual
Coins (投币) and favourites (收藏)	Yes, on every video	Varies	Manual
Pricing	Pay per result, failed lookups free	Subscription + your account	Quota-limited

---

🏆 Top China Scrapers

	RedNote Scraper Notes, user profiles and comments from 小红书
	Douyin Scraper Creator profiles, videos and comments from 抖音
	Weibo Scraper Posts and user profiles from 微博
	Bilibili Scraper  YOU ARE HERE

---

FAQ

What is a Bilibili scraper?

A Bilibili scraper turns a keyword, a video link or a creator's space URL into structured data: the videos with their plays, likes, coins, favourites and comment counts, the comment threads, the danmaku, the subtitles and the UP主 behind them. This Actor returns that as JSON rows, without opening B站 or logging in.

What is danmaku (弹幕)?

Danmaku are the comments that scroll across a Bilibili video as it plays, anchored to the second they were posted at. That anchoring is what makes them useful: this Actor returns each bullet with its progress in seconds, so audience reaction can be mapped to the moment that caused it instead of to the video as a whole.

Do I need a Bilibili account or cookies?

No. There is no login, no session and no cookie, so nothing expires and no account of yours is at risk.

Is there an official Bilibili API?

There is no public international one. Bilibili's developer platform requires Chinese phone verification, and most public data sits behind WBI request signing. This Actor handles that and returns structured JSON with no authentication of yours involved.

Can I use this from ChatGPT or Claude?

Yes, two ways. Paste the "Copy to your AI assistant" block above into any LLM and it will write the call for you. Or connect the Bilibili MCP server config above, and every Atomus Actor becomes a native tool your assistant can call on its own, including chaining several in one request.

What if a video has no subtitles?

You get a row with available: false and an empty segments array, and you are not charged for it. Bilibili's subtitle endpoint is also intermittent — it often answers only on a retry for the same video — so the Actor retries a few times before concluding a video has no track, at no extra cost to you.

How much does it cost?

Per result, per mode: $0.01 per video from search or a creator's catalogue, $0.02 per video detail, $0.008 per comment (minimum 2 per video), $0.012 per UP主 profile, $0.01 per video for danmaku or a transcript, $0.008 per trending result (minimum 2). Gold plans and above pay 20% less. Failed lookups are free.

Can I cap what one run costs?

Yes. Set a per-run charge limit on the Actor's run options: results are delivered and billed page by page, so the run stops at your limit and keeps everything it had already collected. maxItems is the other lever, and the more precise one.

Can I get every video a creator posted?

Yes — user-videos with a space URL or UID walks the creator's whole catalogue up to your maxItems, billed at the same $0.01 per video as keyword search.

Is it legal to scrape Bilibili?

This Actor reads publicly visible content, the same data any visitor sees without logging in. You are responsible for using the output in line with applicable laws (GDPR/CCPA), Bilibili's terms, and your own compliance requirements. It is an independent tool, not affiliated with Bilibili.

Can I run this on a schedule?

Yes. Apify Schedules run the Actor on a cron interval and webhooks push each finished run into your systems — which is how a keyword becomes a trend monitor.

---

哔哩哔哩采集器中文说明

无需 Cookie、无需登录，按关键词或链接抓取 B站（哔哩哔哩）数据：视频搜索、视频详情、评论、弹幕、原生字幕转写、UP主主页、UP主投稿视频、热门与热搜，共八种模式。每个关键词或目标最多 5,000 条，按结果计费，查询失败不收费。弹幕（每条带精确到秒的出现时间）与 B站原生字幕轨道（带时间轴的分段 + 全文），是本采集器与普通视频采集器最大的不同。

八种抓取模式

用 searchType 选择模式，然后只填写该模式读取的字段。输入 JSON 与上方英文部分完全一致。

模式（searchType）	输入	返回
关键词搜索（search）	关键词	匹配关键词的视频，含播放、点赞、投币、收藏、弹幕与评论数
视频详情（video-detail）	视频链接或 BV 号	单个视频的完整数据，含 cid、分P 列表、清晰度、原创/转载
视频评论（comments）	视频链接或 BV 号	评论及楼中楼，含 IP 属地、等级、大会员、认证、UP主是否点赞或回复
弹幕（danmaku）	视频链接或 BV 号	该视频的弹幕，每条带在视频中出现的秒数
字幕/转写（transcript）	视频链接或 BV 号	原生字幕轨道（含自动生成的 ai-zh）的时间轴分段与全文
UP主主页（profile）	主页链接或 UID	UP主资料卡：粉丝、关注、等级、认证、大会员
UP主投稿视频（user-videos）	主页链接或 UID	该 UP主的全部投稿视频
热门/热搜（trend）	无需输入	热门视频（含上榜理由）或热搜词（含排名与热度）

输入参数

参数	类型	默认值	说明
searchType	string	search	抓取模式，必填：search、video-detail、comments、danmaku、transcript、profile、user-videos、trend。
keywords	string[]	无	搜索关键词，用于「关键词搜索」。
videoUrls	string[]	无	视频链接或 BV 号，用于「视频详情」「评论」「弹幕」「字幕」。
userUrls	string[]	无	UP主主页链接或纯数字 UID，用于「UP主主页」「UP主投稿视频」。
maxItems	integer	20	每个关键词或目标的最大结果数，上限 5000；用于搜索、评论、UP主投稿与热门。
sortOrder	string	totalrank	免费筛选，仅「关键词搜索」：totalrank 综合、click 最多播放、pubdate 最新、dm 最多弹幕、stow 最多收藏。
durationFilter	string	any	免费筛选，仅「关键词搜索」：short（10分钟以下）、medium（10-30）、long（30-60）、verylong（60分钟以上）。
publishedAfter / publishedBefore	string	无	免费筛选，仅「关键词搜索」：YYYY-MM-DD 日期范围。
trendType	string	videos	仅「热门」：videos 热门视频 或 keywords 热搜词。
includeReplies	boolean	false	💲 仅「视频评论」：抓取每条评论下的全部楼中楼，每条回复按一条评论计费，并计入 maxItems。

支持的链接格式：视频为 https://www.bilibili.com/video/BV1rpWjevEip 或纯 BV 号 BV1rpWjevEip；UP主为 https://space.bilibili.com/178360345 或纯数字 UID 178360345。其他格式会被跳过并记录日志，不计费。同一次输入中重复的关键词、链接或 UID 会先去重，因此重复粘贴不会重复扣费。maxItems 是列表类模式的主要费用控制项：总费用≈模式单价×结果数×目标数；「视频详情」「UP主主页」每个目标一条，「弹幕」「字幕」按单次调用返回全部并按视频计费。

输出字段

每条结果为数据集中的一行，均带有 searchType 标明来源模式；列表类结果另带来源标识（keyword、video、uid、trendType）。

• 视频（搜索 / 详情 / UP主投稿 / 热门）：bvid、aid、title、desc、url、cover、duration（秒）、category（分区）、tags[]、pubdate、part_count，互动数据 view_count、like_count、coin_count（投币）、favorite_count（收藏）、share_count、danmaku_count、reply_count，以及嵌套的 owner（mid、name、avatar）。仅「视频详情」另有 cid、parts[]（每个分P 的 cid、page、title、duration）、dimension{width,height,rotate}、copyright（1 原创 / 2 转载）、dynamic（UP主动态文案）；仅「热门」另有 rcmd_reason（上榜理由）与 pub_location（发布地）。
• 评论：rpid、content、like_count、reply_count、ctime、root、oid、location（归属地）、up_liked、up_replied、user_level、user_vip、user_verified、user（mid、name、avatar），以及 replies[]（结构相同，嵌套）。
• 弹幕：dmid、text、progress（视频内秒数）、mode（1-3 滚动、4 底部、5 顶部）、fontsize、color（十进制 RGB）、ctime、sender_hash。
• 字幕/转写：available、lang（如 ai-zh）、ai、tracks[]{lan,lan_doc,ai,url}、segments[]{from,to,content}、text（全文）。
• UP主主页：mid、name、avatar、sign、level、fans、following、friends、place、description、verified、verify_desc、vip、url。
• 热搜词：rank、keyword、show_name、heat_score、icon。

cid 只出现在「视频详情」的结果中；弹幕与字幕模式会自动解析 cid，不额外收费。示例结果见上方英文部分的 JSON。

价格

按结果计费，无订阅。每个模式是独立的计费事件，Apify Gold 及以上套餐享 8 折。

模式	事件	Free–Silver	Gold+
关键词搜索	video-scraped	$0.01 / 视频	$0.008
UP主投稿视频	video-scraped	$0.01 / 视频	$0.008
视频详情	video-detail-scraped	$0.02 / 视频	$0.016
评论（含每条回复）	comment-scraped	$0.008 / 条	$0.0064
UP主主页	profile-scraped	$0.012 / 个	$0.0096
弹幕	danmaku-scraped	$0.01 / 视频	$0.008
字幕/转写	transcript-scraped	$0.01 / 视频	$0.008
热门 / 热搜	trend-scraped	$0.008 / 条	$0.0064

最低计费：评论每个视频最低 2 条，热门每次运行最低 2 条。为什么有最低计费：一次请求就是一次查询，无论返回 50 条、1 条还是 0 条，工作量都相同。最低计费覆盖的是这次请求本身，这样结果很少的目标依然值得抓取，而不是被直接拒绝。

免费的部分：全部搜索筛选（sortOrder、durationFilter、publishedAfter、publishedBefore）；每条评论最多 3 条楼中楼预览；弹幕与字幕所需的 cid 解析；跨页去重，重复的视频不会重复计费；没有字幕轨道的视频（返回 available: false 的一行且不计费）；以及失败的请求，永不计费。

免费套餐：每个自然月 10 条搜索结果，外加其他每个模式各 1 条，足以在付费前查看所有字段结构；「关键词搜索」与「UP主投稿视频」共用这 10 条额度。计数在每月 1 日重置，任何付费 Apify 套餐都会解除该上限。

常见问题

什么是弹幕？ 弹幕是随视频播放滚动而过的即时评论，锚定在其发布的那一秒。本采集器返回每条弹幕的 progress（视频内秒数），因此可以把观众反应对应到引发它的具体画面，而不只是整个视频。

需要 Cookie 或账号吗？ 不需要。没有登录、没有会话、没有 Cookie，因此不会过期，也不涉及你的任何账号风险。

有官方 API 吗？ 没有面向国际开发者的公开 API：B站开发者平台需要中国手机号认证，且大部分公开数据受 WBI 签名保护。本采集器已处理这些细节，无需你提供任何认证。

可以在 ChatGPT 或 Claude 里使用吗？ 可以。把上方「Copy to your AI assistant」代码块粘贴给任意 LLM，它就能替你写好调用；或者配置上方的 Bilibili MCP 服务器，让 Atomus 的每个 Actor 成为智能体可直接调用的工具。

视频没有字幕怎么办？ 返回 available: false 且 segments 为空的一行，不计费。B站字幕接口本身不稳定（同一视频常常要重试才返回），因此本采集器会先重试若干次再判定为没有字幕，这些重试不额外收费。

可以限制单次运行的花费吗？ 可以。在运行设置里设置单次运行的计费上限即可：结果是逐页交付并逐页计费的，达到上限时运行会停止，并保留已经采集到的全部结果。maxItems 是更精确的另一个控制项。

可以定时运行吗？ 可以。用 Apify Schedules 按 cron 周期运行，用 webhooks 把每次运行结果推送到你的系统，关键词监控就是这样搭起来的。

抓取哔哩哔哩合法吗？ 本采集器仅访问公开可见的数据（无需登录即可查看的数据），不绕过认证，不访问私密数据。请自行确保使用方式符合适用法律（GDPR、CCPA）、哔哩哔哩的服务条款以及你自身的合规要求。

免责声明

本采集器为独立工具，与哔哩哔哩（Bilibili Inc. / 上海宽娱数码科技有限公司）无隶属、背书或赞助关系。「哔哩哔哩」与「Bilibili」为其各自所有者的商标。请在遵守适用的数据保护法律（GDPR、CCPA）与哔哩哔哩服务条款的前提下使用本采集器输出的数据，不得用于垃圾信息、骚扰或任何违法用途。

---

All Atomus scrapers

2.4M+ RESULTS DELIVERED

LinkedIn	Profile · Posts · Reactions · Comments · Company · Employees
Lead gen	Leads Finder
Google Maps	Places & local businesses
TikTok	Videos · Comments
X (Twitter)	Tweets & profiles
RedNote 小红书	Notes, users & comments
Douyin 抖音	Profiles, videos & comments
Weibo 微博	Posts & profiles
Bilibili 哔哩哔哩	Videos & creators

---

Support

Hey, I'm Chico, founder of Atomus. I built this Actor and I answer the messages about it. Something broke? A field you need isn't there? Not sure it fits what you're doing? Send me a message, most answers come the same day. 💬  DM me on LinkedIn    or hello@dendelabs.com

---

⚠️ Disclaimer

This Actor is an independent tool and is not affiliated with, endorsed by, or sponsored by Bilibili (上海宽娱数码科技有限公司 / Bilibili Inc.). "Bilibili" and "哔哩哔哩" are trademarks of their respective owner. All trademarks are property of their respective owners.

Use the data extracted by this Actor in compliance with applicable data protection laws (GDPR, CCPA) and Bilibili's terms of service. Do not use it for spam, harassment, or unlawful purposes.