Bilibili Search Scraper — Full Stats · No Login · $0.01

Search Bilibili, China's largest video and streaming platform, by keyword and get back one record per video with full engagement stats: views, danmaku, likes, favorites, comments, duration, cover image, uploader and tags. No Bilibili account, no login and no cookies required.

Why use this scraper

Bilibili has no public search API for third parties, and its engagement metrics (danmaku count, favorites, coins) are unique to the platform and not replicated by any general video search tool. Brands running Chinese social listening, researchers tracking gaming and anime content trends, and creators scouting competing channels all need these numbers, but pulling them by hand from the site does not scale past a handful of searches.

This actor reads the same public search results Bilibili serves to any visitor, with no login wall in the way, and returns a clean, structured record per video including every stat Bilibili shows on the result card. Search by any keyword, in English or Chinese, and sort the results the way Bilibili itself allows.

Overview

Input is a search keyword plus an optional sort order. The actor queries Bilibili's public search and returns one JSON record per video: identifiers (bvid, aid), title, description, duration, publish date, cover image, category, uploader name and profile URL, and the engagement numbers (views, danmaku, likes, favorites, comments) and tags. A search for 30 videos typically finishes in under a minute.

Supported inputs

• Search keyword (keyword) — any search term, in English or Chinese, e.g. "genshin impact" or "美食".
• Sort order (sortBy) — relevance (default), click (most viewed), pubdate (newest), danmaku (most danmaku), favorites (most favorited) or comments (most commented).
• Max videos (maxItems) — caps how many videos are returned for the search (default 30, up to 300).
• Proxy configuration — Apify Proxy is enabled by default and recommended, since Bilibili rate-limits by IP; most users can leave this at its default.

Use cases

• Social listening and brand monitoring teams — track how a brand, game or show is being covered on Bilibili by searching its name and sorting by newest or most viewed.
• Gaming and anime marketing teams — scout which creators are already covering a title and how much engagement (danmaku, favorites) their videos draw.
• Market researchers — measure relative interest in a topic on Bilibili by comparing view counts and danmaku density across search results.
• Competitive content research — pull a competitor channel's best-performing videos on a topic by keyword and sort order, without opening Bilibili manually.
• Trend and content-calendar teams — run a recurring search on a topic to spot emerging videos before they peak in views.

How it works

The actor queries the same public search endpoint Bilibili's own search page uses, with no login required, and reads each result's metadata and engagement stats directly from that response. Because this is the public search surface rather than a logged-in API, there is no Bilibili account on your side that can be rate-limited or banned; the built-in proxy support exists purely to spread request volume and avoid IP-based throttling on larger searches.

Input configuration

Simple keyword search:

{
  "keyword": "genshin impact",
  "maxItems": 30
}

Most-viewed videos on a topic:

{
  "keyword": "美食",
  "sortBy": "click",
  "maxItems": 50
}

Newest videos with heavy proxy use for a larger pull:

{
  "keyword": "unboxing",
  "sortBy": "pubdate",
  "maxItems": 200,
  "proxyConfiguration": { "useApifyProxy": true }
}

Output sample

Real output from a live run:

[
  {
    "bvid": "BV1zqW765EJ9",
    "aid": 117579762961393,
    "title": "Genshin Impact 7.3 Weekly Boss Gameplay Full Walkthrough Demo",
    "url": "https://www.bilibili.com/video/BV1zqW765EJ9",
    "description": "Check out the new Version 7.3 Weekly Boss gameplay in Genshin Impact!...",
    "durationText": "4:7",
    "publishedAt": "2026-11-08T08:07:46.000Z",
    "coverUrl": "https://i1.hdslb.com/bfs/archive/3292e3b969fe1b65598f70e4747ddeafec0a16cc.jpg",
    "category": "手机游戏",
    "keyword": "genshin impact",
    "authorName": "tsaritsadaily",
    "authorMid": 1551566300,
    "authorUrl": "https://space.bilibili.com/1551566300",
    "views": 18621,
    "danmaku": 100,
    "likes": 193,
    "favorites": 124,
    "comments": 140,
    "tags": ["genshin", "genshin impact", "genshin 7.3", "genshin news"],
    "scrapedAt": "2026-11-09T09:18:35.670Z"
  }
]

Key output fields

Field	Type	Description
bvid	string	Bilibili's video identifier (the one shown in the video URL)
aid	number	Bilibili's numeric legacy video ID
title	string	Video title
url	string	Canonical Bilibili video URL
description	string	Video description as written by the uploader
durationText	string	Video duration as shown on the result card
publishedAt	string	ISO publish timestamp
coverUrl	string	Thumbnail/cover image URL
category	string	Bilibili's content category (often in Chinese)
keyword	string	The search keyword that returned this video
authorName	string	Uploader's display name
authorMid	number	Uploader's numeric Bilibili user ID
authorUrl	string	Uploader's channel URL
views	number	View count at scrape time
danmaku	number	Danmaku (floating comment) count at scrape time
likes	number	Like count at scrape time
favorites	number	Favorite (collection) count at scrape time
comments	number	Comment count at scrape time
tags	array of strings	Tags attached to the video by its uploader
scrapedAt	string	ISO timestamp of the scrape

Pricing

Pay-per-event, charged only for videos actually returned:

Event	Price
Video scraped	$0.01 per video

At $0.01/video, 1,000 videos cost $10.00. One flat price for the full stat set above, no separate charge for any of the included fields.

What makes this richer than the competition

zhorex/bilibili-scraper, the highest-traffic Bilibili actor on the Store, charges $0.02/item, double this actor's $0.01, and additionally charges $0.05 per video as a separate "danmaku reaction profile" event if a buyer wants danmaku detail beyond a raw count. This actor includes the raw danmaku count in every video record at no extra charge, so a buyer who just needs the danmaku volume alongside the other stats does not need to opt into and pay for a second event type. lofomachines/bilibili-scraper, the second-highest-traffic actor in the category, covers a broader feature set (profiles, comments, search) across multiple actors or modes; this actor stays scoped to search results with full stats in a single call, which keeps the input simple for the common case of "search a keyword, get ranked videos with their numbers."

Notes & limitations

• description can be truncated by Bilibili itself on the search result card rather than by this actor; the full description is only visible on the video's own page.
• category and some text fields come back in Chinese even for an English-language search keyword, reflecting how Bilibili itself labels the content; this is not translated by the actor.
• The actor searches and returns videos only; it does not fetch comments, danmaku text content or uploader channel statistics beyond what is shown on the search card.
• Engagement numbers (views, likes, favorites, comments, danmaku) reflect the moment of the scrape and will differ from a later check, since they change continuously on Bilibili.
• A very niche keyword with few matching videos on Bilibili returns fewer items than maxItems; this reflects actual search volume, not a scraping shortfall.
• Live streams and some member-exclusive content are not indexed by Bilibili's public search and will not appear in results.

FAQ

Does this need a Bilibili account or login? No. The actor queries the same public search any visitor can use without signing in.

Can I search with a Chinese keyword? Yes, the actor accepts any keyword in any language Bilibili's search supports, including Chinese.

Does it return the actual danmaku comments, or just the count? Just the count. This actor returns the danmaku number shown on the search card; scraping the individual danmaku comment text is out of scope.

Can I sort by most-commented instead of most-viewed? Yes, set sortBy to comments for that order, or click for most-viewed, favorites for most-favorited, and danmaku for most danmaku.

Does it work for live-stream search? No, this actor is scoped to regular uploaded videos as returned by Bilibili's video search, not live streams.

How many videos can I pull in one run? Up to 300 per run via maxItems; for more, run the actor again with a different keyword or sort order to cover a broader slice of the topic.

Does the price change based on my Apify plan? No. This actor charges a flat $0.01 per video regardless of your Apify plan tier.

For AI Agents & LLM Apps

This actor is available over the Apify MCP server (mcp.apify.com) for direct use from Claude, Cursor or any MCP-compatible agent, and over the plain REST API for scripted pipelines.

Synchronous call that returns dataset items directly, no polling required:

curl "https://api.apify.com/v2/acts/gio21~bilibili-scraper/run-sync-get-dataset-items?token=<APIFY_TOKEN>" \
  -X POST \
  -H "Content-Type: application/json" \
  -d '{"keyword": "genshin impact", "maxItems": 30}'

A single keyword with maxItems around 30 to 50 is the sweet spot for most social-listening checks, keeping runtime short while still covering a representative slice of a topic's coverage on Bilibili.

Disclaimer

This actor is an independent tool that extracts publicly available Bilibili search data. It is not affiliated with, endorsed by or sponsored by Bilibili Inc. "Bilibili" is a trademark of its respective owner. Use it in line with Bilibili's own terms and any applicable data protection law in your jurisdiction.

SEO Keywords

Bilibili scraper, Bilibili search API, Bilibili video data, Bilibili danmaku scraper, Chinese video platform scraper, no-login Bilibili scraper.