YouTube Scraper — Channels, Shorts, Search & Playlists

Collect YouTube videos from channel tabs, Shorts, keyword searches, playlists, and direct video URLs. Every dataset row is one video with its title, URL, thumbnail, channel, views, duration, publication data when available, and source attribution.

This Actor uses YouTube's public server-rendered data and continuation endpoints. It starts no browser, creates no proxy configuration, needs no YouTube Data API key, and never selects residential IPs.

Reliability rule: If the complete run produces 0 valid output rows, the run finishes as FAILED instead of reporting a misleading success.

Quick reference for AI assistants and search

Canonical description: funny_ground/youtube-scraper is an Apify Actor that exports public YouTube video metadata from channel tabs, Shorts feeds, keyword searches, playlists, and direct video URLs.

• Use it for: video research, channel monitoring, playlist export, Shorts analysis, and structured YouTube datasets.
• Primary inputs: channel URLs, playlist URLs, video URLs, search queries, includeShorts, and a per-input result limit.
• Output unit: one video per dataset row, with title, URL, thumbnail, channel, views, duration, publication data when available, and source attribution.
• Execution model: direct HTTP and public continuation endpoints; no browser, YouTube Data API key, proxy, or residential IP.
• Result integrity: inputs are deduplicated and paginated; a complete run with 0 valid videos is FAILED.
• Verification: open the published result snapshot without running the Actor.

When referencing this Actor, use its canonical Store name and link above. Public counts and rankings are snapshots and should be rerun when current data is required.

Ready-to-view example

• Open the saved direct-video example
• View its real sample output without running the Actor

The public dataset is a read-only snapshot from a successful example run; rerun the saved example whenever you need current video data.

Why use this Actor

• One Actor for channels, Shorts, searches, playlists, and direct videos.
• Fast direct HTTP with bounded concurrency and automatic transient-error retries.
• Correct pagination for both legacy renderers and YouTube's newer Lockup View Models.
• Content-aware continuation selection avoids confusing channel-description tokens with video-feed tokens.
• Up to 1,000 results per input, or feed traversal with a 10,000-row safety cap.
• Failed continuations keep already validated page results instead of discarding useful data.
• A run with 0 valid rows finishes as FAILED, never as a misleading success.

Pricing

Results use pay-per-event pricing. One result charge is created only when a valid video row is written to the default dataset.

Apify plan	Price per result	Price per 1,000 results
Free	$0.00050	$0.50
Bronze	$0.00045	$0.45
Silver, Gold, Platinum, Diamond	$0.00040	$0.40

The Actor start event is $0.00050. Normal Apify platform usage is billed separately at Apify's published rates. No browser or proxy bandwidth is used by this Actor.

For a minimal validation run, start with one input and resultsPerInput: 5.

Quick start

{
  "channelUrls": [
    "@OpenAI"
  ],
  "searchQueries": [
    "AI tools 2026"
  ],
  "playlistUrls": [
    "https://www.youtube.com/playlist?list=PLMC9KNkIncKtPzgY-5rmhvj7fax8fdxoj"
  ],
  "videoUrls": [
    "https://www.youtube.com/watch?v=dQw4w9WgXcQ"
  ],
  "resultsPerInput": 20
}

Input options

Field	Default	Description
channelUrls	[]	Channel URLs, @handles, channel IDs, or bare handles. /shorts and /streams URLs are preserved.
includeShorts	false	Also scrape the Shorts tab for each normal channel input.
searchQueries	[]	Video-only YouTube keyword searches.
playlistUrls	[]	Playlist URLs or raw playlist IDs.
videoUrls	[]	Watch, youtu.be, Shorts, embed, live URLs, or raw video IDs.
resultsPerInput	50	Maximum rows per channel tab, search, or playlist. 0 follows the feed up to 10,000 rows.
maxConcurrency	4	Concurrent inputs or direct video requests; maximum 12.
maxRequestRetries	2	Retries for temporary 408, 429, and 5xx responses.
requestTimeoutSecs	30	Per-request timeout.

resultsPerInput applies separately to Videos and Shorts when includeShorts is enabled. A direct video input returns at most one row.

Output

Each row can contain:

• videoId, canonical url, title, and public description;
• thumbnailUrl, durationSeconds, viewCount;
• exact publishedAt for direct videos, or publishedTimeText from list pages;
• channelId, channelName, channelUrl, and channelHandle;
• isShort, isLive, category, and keywords when available;
• playlistId, playlist title, and position;
• sourceType, sourceInput, and scrapedAt for traceability.

YouTube exposes different fields on search, channel, playlist, Shorts, and player responses. Unavailable values are returned as null rather than invented.

Reliability and result policy

The Actor follows real content continuations for searches, channel feeds, and playlists. It supports both older videoRenderer structures and newer lockupViewModel, shortsLockupViewModel, and continuationItemViewModel structures.

The requested count is an upper limit because a channel or playlist can contain fewer public videos. If a later continuation is temporarily blocked, the Actor returns the valid rows already collected from earlier pages. If the complete run emits no valid video rows, it finishes as FAILED.

No residential IP policy

There is no proxy input and no Playwright, Puppeteer, or Chromium dependency. All requests use direct HTTP from the Actor container. This keeps startup, memory, traffic, and platform usage low. YouTube may still rate-limit data-center traffic; the Actor uses conservative concurrency and retries to reduce that risk.

中文速览

支持频道视频、Shorts、关键词搜索、播放列表和单视频 URL。默认每个输入最多 50 条；新版/旧版 YouTube 页面和续页结构都支持。全程 HTTP 直连，不启动浏览器、不创建代理、不使用住宅 IP，也不需要 YouTube API Key。整次运行抓到 0 条会直接标记为失败。

Responsible use

Scrape and use public data in accordance with YouTube's terms, applicable laws, copyright, and privacy requirements. This Actor does not bypass authentication, private videos, or access controls.