YouTube Video Comments Scraper

Scrape YouTube video comments (and optionally their replies) by video ID. Returns clean, structured data for each comment: text, author channel, like count, reply count, published time, verified / creator flags, and the avatar URL.

Unofficial. This Actor is not affiliated with, authorized, or endorsed by YouTube or Google. It is an independent tool that retrieves publicly available data via a third-party API. Use it in compliance with YouTube's terms of service and all applicable laws; you are responsible for how you use the retrieved data.

What it does

• Comments by video — give one or more YouTube video IDs; the Actor paginates through the video's comments automatically. Sort by top or newest.
• Replies (optional) — for each top-level comment, fetch the replies underneath it, capped per comment.
• De-duplicates comment IDs within a run and stops at maxItems.

Input

Field	Type	Default	Description
videoId	string	dQw4w9WgXcQ	The 11-char YouTube video ID (after watch?v=).
videoIds	string[]	[]	Optional extra video IDs to scrape in the same run.
maxItems	integer	20	Max total comments + replies across all videos.
sortBy	enum	top	top or newest.
includeReplies	boolean	false	Also fetch replies under each comment.
maxRepliesPerComment	integer	10	Cap on replies per comment when enabled.
languageCode	string	en	YouTube interface language (hl).
countryCode	string	US	YouTube content region (gl).

Example input

{
  "videoId": "dQw4w9WgXcQ",
  "maxItems": 200,
  "sortBy": "top",
  "includeReplies": true,
  "maxRepliesPerComment": 5
}

Output

One dataset item per comment (or reply):

{
  "commentId": "Ugzge340dBgB75hWBm54AaABAg",
  "videoId": "dQw4w9WgXcQ",
  "parentCommentId": null,
  "content": "can confirm: he never gave us up",
  "publishedTime": "1 year ago",
  "likeCount": 240000,
  "likeCountText": "240K",
  "replyCount": 962,
  "replyLevel": 0,
  "isReply": false,
  "author": {
    "channelId": "UCBR8-60-B28hp2BmDPdntcQ",
    "displayName": "@YouTube",
    "channelUrl": "https://www.youtube.com/@YouTube",
    "avatarUrl": "https://yt3.ggpht.com/...",
    "isVerified": true,
    "isCreator": false,
    "isArtist": false
  },
  "creatorThumbnailUrl": "https://yt3.ggpht.com/...",
  "replyContinuationToken": "Eg0SC2RR...",
  "source": "video:dQw4w9WgXcQ"
}

Replies carry isReply: true, a non-null parentCommentId, and source of the form replies:<parentCommentId>.

Notes

• Data is sourced live; YouTube occasionally rate-limits, so the Actor retries transient blocks with exponential backoff.
• Like / reply counts are normalized to integers where the source provides a parseable value (e.g. 240K, 24万 → 240000); the original display string is kept in likeCountText.

Quick start

1. Open the Actor and press Run — the default input works out of the box.
2. Adjust the input fields below to your target (keywords, IDs, or URLs) and set maxItems to cap spend.
3. Grab results from the Dataset tab as JSON / CSV / Excel, or pull them via the Apify API and MCP from your own code.

No proxies to configure, no cookies to paste, no login — the Actor handles everything server-side.

Why teams switch to this YouTube comments scraper

YouTube comments are rich audience feedback, but the official API rate-limits them and browser scrapers stall on long threads. This Actor walks a video's comments via a direct HTTP API and returns them with author, likes, and reply data as clean JSON at $3 per 1,000 comments.

What people build with it

• Sentiment analysis — turn a video's comments into a structured readout of how the content landed.
• Voice-of-customer mining — feedback and questions under review and tutorial videos, verbatim.
• Community management — collect comments across your channel daily and prioritise replies.
• Creator research — study how top channels' audiences react to understand what resonates.
• Conversation datasets — author-linked comment threads for fine-tuning.

Tips for better results

• Accepts video URLs or bare video IDs.
• Find comment-heavy videos first with YouTube Search Scraper.
• Pair with YouTube Transcript Scraper to analyse what was said and how viewers reacted, side by side.

---

Why this Actor

• Direct API, no headless browser — fast, stable runs with nothing to babysit.
• No login, no cookies — we never touch your accounts, so there's no ban risk.
• Fresh, real-time data — every run reads the source live, not a stale cache.
• Pay per result — you're billed only for the rows actually delivered.
• Structured JSON — export to CSV, Excel, or JSON, or pull straight from the API / MCP.

Use cases

• Mine audience sentiment and feature requests from real comment threads.
• Surface the most-liked replies and frequent questions under any post.
• Build moderation, UGC, or social-listening datasets at scale.
• Spot superfans and detractors by author and engagement.

FAQ

Do I need an account, cookies, or to log in anywhere? No. The Actor talks to a fast, direct HTTP API server-side — you just provide inputs and run it.

How am I billed? Pay-per-result: a fixed price per row returned, with no separate platform/compute charge. Caps like maxItems keep spend predictable.

Can I run it on a schedule or call it from my app? Yes — use Apify Schedules, the REST API, the JavaScript / Python clients, or the MCP server. See the API tab.

Is this affiliated with YouTube? No. It's an independent tool that collects publicly available data. Use it in line with the platform's terms and applicable law.

More YouTube scrapers by us

• YouTube Transcripts — Subtitles / transcripts for LLM datasets
• YouTube Search — Keyword video search · stats · channels
• YouTube Channel Videos — All videos for a channel · stats
• YouTube Channel Info — Channel profile · subs · about

Browse the full fleet → https://apify.com/ethereal_wool