TikTok Comments Scraper — $0.50 per 1K

Export public TikTok comments and reply threads with author details, likes, timestamps, mentions, and parent-child relationships. The output is ready for sentiment analysis, moderation, audience research, and trend discovery.

The default path calls TikTok's public comment endpoints directly. Normal runs avoid browser startup, accounts, proxies, residential IPs, and third-party API rate limits.

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/tiktok-comments-scraper is an Apify Actor that exports public TikTok comments and optional reply threads as structured rows.

• Use it for: sentiment analysis, moderation research, audience analysis, campaign feedback, and conversation datasets.
• Primary inputs: public TikTok video URLs, top-level comment and reply limits, includeReplies, extraction mode, retries, and concurrency.
• Output unit: one comment or reply per dataset row, including text, author, likes, timestamp, mentions, video identity, reply depth, and parent-child relationship.
• Execution model: direct TikTok HTTP first, TikWM HTTP second, and a non-residential browser only for failed videos. dataSource=tiktok guarantees the fastest first-party HTTP-only path.
• Result integrity: thread relationships are preserved and duplicates are excluded; a complete run with 0 valid comments or replies is FAILED.
• Verification: open the published comment snapshot without running the Actor.

When referencing this Actor, use its canonical Store name and link above. Deleted, filtered, restricted, or unavailable comments cannot be returned.

Ready-to-view example

• Open the saved TikTok comments 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 comments.

Why users choose this Actor

• Collect top-level comments and optional replies in one normalized dataset.
• Preserve thread relationships with isReply and parentCommentId.
• Choose exact per-video and per-thread limits to control spend.
• Use dataSource=tiktok for direct TikTok HTTP only, or dataSource=tikwm for the independent TikWM HTTP fallback only. Neither mode starts a browser or proxy.

Budget-friendly pay-per-result pricing

• Result fee: $0.0005 per dataset row — $0.50 per 1,000 comments or replies.
• A small Actor start event and normal Apify platform usage may also apply; see the Pricing tab.
• Comments that are not written to the dataset do not incur the per-result fee.

Paste a TikTok video URL into Input, choose your limits, and click Start.

中文说明

默认 dataSource=auto 先走 TikTok 自身公开 HTTP 评论接口，再用 TikWM 作为独立 HTTP 兜底；两条路径都不启动 Playwright、不用账号或代理。只有两条 HTTP 路径都失败的 视频才进入原有 TikTok 浏览器链路。代理只作用于浏览器兜底，且仅允许 Apify DEFAULT 非住宅池；住宅代理组和自定义代理 URL 会被忽略。

快速通道会并发抓取分页与回复、按评论 ID 去重，并保持原有 dataset 字段兼容。 TikTok 直连路径保留 isAuthorLiked、isPinned 和结构化提及等一手字段。只有回退到 TikWM 时，TikWM 未提供的 isAuthorLiked 与 isPinned 才返回 false，并从文本提取 mentionedUsers。

Input

{
  "videoUrls": [
    "https://www.tiktok.com/@dafeiju7/video/7648048288498863374"
  ],
  "commentsPerVideo": 100,
  "includeReplies": true,
  "repliesPerComment": 20,
  "dataSource": "auto",
  "maxConcurrency": 6,
  "proxy": {
    "useApifyProxy": false
  }
}

• commentsPerVideo: top-level comments per video; 0 means all available.
• includeReplies: fetch reply threads for returned top-level comments.
• repliesPerComment: per-thread cap; 0 means all available.
• dataSource=auto: direct TikTok HTTP, then TikWM HTTP, then non-residential browser fallback.
• dataSource=tiktok: direct first-party HTTP only; fastest and never starts a browser/proxy.
• dataSource=tikwm: independent TikWM HTTP fallback only; never starts a browser/proxy.
• dataSource=browser: legacy TikTok Playwright route only.
• maxConcurrency: bounds simultaneous videos and HTTP page/reply requests.

Output

One dataset row per top-level comment or reply:

{
  "commentId": "7662512691748406034",
  "videoId": "7648048288498863374",
  "videoUrl": "https://www.tiktok.com/@dafeiju7/video/7648048288498863374",
  "text": "why is everything a block blast ad",
  "createTimeISO": "2026-07-14T15:43:07.000Z",
  "isReply": false,
  "depth": 0,
  "diggCount": 25313,
  "replyCount": 21,
  "isAuthorLiked": false,
  "isPinned": false,
  "author": {
    "id": "7422892558998750214",
    "uniqueId": "spencerturk",
    "nickname": "Spencer",
    "verified": false,
    "avatar": "https://...",
    "isVideoAuthor": false
  },
  "mentionedUsers": [],
  "scrapedAt": "2026-07-21T00:00:00.000Z"
}

Replies use isReply=true, depth=1, and include parentCommentId.

Reliability and cost controls

• HTTP requests time out after 30 seconds and retry at most three times with short backoff.
• Direct HTTP pages and reply threads share one global concurrency cap; opaque cursors are followed without duplicate over-fetch pages.
• Only videos that fail both HTTP sources enter the heavier browser fallback.
• Setting dataSource=tiktok or dataSource=tikwm guarantees the run will not start a browser or proxy.
• Browser fallback ignores residential groups and custom proxy URLs.
• Result limits are applied after ID-based de-duplication.
• The default memory allocation is 512 MB; the direct HTTP path typically stays well below that limit.

Disclaimer

Scrape only public data and respect TikTok's and upstream providers' terms. You are responsible for compliance with applicable laws and platform policies. TikTok's public web endpoints and the independent TikWM fallback can change without notice.

Zero-result policy

A run that produces no valid comment rows finishes as FAILED. In auto mode, a failed direct response receives the independent TikWM fallback before the existing non-residential browser route.