YouTube Comments API Actor

This Apify actor extracts comments from public YouTube videos without requiring an API key. It uses a direct scraping approach to bypass YouTube API quotas and provides structured data suitable for analysis.

Overview

The actor downloads comments from any public YouTube video. You provide video URLs or IDs, and it returns the comments in a structured JSON format along with video metadata. It handles multiple languages, offers sorting options, and is built to process large numbers of comments efficiently.

Key Features

• No API Key Required: Operates via direct scraping, avoiding YouTube API limits.
• Flexible Input: Accepts an array of YouTube video URLs or video IDs.
• Configurable Output: Control the number of comments, sort order (top or newest), and language.
• Structured Data: Returns rich metadata for each comment (author, likes, reply count, timestamps) and a summary for each video.
• Built-in Error Handling: Manages common issues like unavailable videos, disabled comments, or invalid URLs gracefully.

How to Use

Configure the actor run with an input JSON object. The main required parameter is urlOrVideoId, which is an array containing one or more video identifiers.

Input Schema

Input Example

{
  "urlOrVideoId": ["https://www.youtube.com/watch?v=mnF3sUfyNRI"],
  "language": "en",
  "max_comments": 100,
  "sortBy": "newest"
}

Output

The actor outputs a dataset where each item corresponds to one processed video. The output contains the video info, the settings used, an array of comments, and a summary.

Output Structure

{
  "video_info": {
    "video_id": "mnF3sUfyNRI",
    "video_url": "https://www.youtube.com/watch?v=mnF3sUfyNRI",
    "download_date": "2025-09-03T20:22:16.881390",
    "downloader": "youtube-comments-api"
  },
  "settings": {
    "max_comments": 100,
    "sort_by": "newest",
    "language": "en",
    "total_downloaded": 95
  },
  "comments": [
    {
      "comment_id": "Ugz_xHS6GebfPxWhAjt4AaABAg",
      "author": "@sandro6411",
      "author_channel_id": "UCkAr1IWgyEBfp49BamNnVTw",
      "text": "what's up with speed's voice ?",
      "time_text": "1 minute ago",
      "time_parsed": 1756930876.09903,
      "like_count": "0",
      "reply_count": 0,
      "is_reply": false,
      "parent_id": null,
      "is_hearted": false,
      "is_pinned": false,
      "is_verified": false
    }
    // ... more comments
  ],
  "summary": {
    "total_comments": 95,
    "main_comments": 90,
    "replies": 5,
    "hearted_comments": 1,
    "pinned_comments": 0
  }
}

If a video cannot be processed (e.g., unavailable, comments disabled), an error object is returned for that specific video, allowing the rest of the batch to complete.

Performance & Best Practices

• Batch Size: Process 10-50 videos per run for optimal performance.
• Resource Allocation: Allocate at least 512MB of memory for large comment sets.
• Timeout: Set the actor timeout to 15-30 minutes when processing many videos or a high max_comments value.
• Error Handling: Check the output for error objects on individual video failures.