TikTok Comment Scraper

Scrape TikTok comments without touching the app

Comments are the densest signal layer on TikTok. They carry sentiment, purchase intent, complaints, language and region hints, and the social proof - likes and reply counts - that decides which opinions surface. For social listening platforms, CX teams, brand-safety tooling and academic research, the comment thread is frequently worth more than the video attached to it.

TikLiveAPI exposes two endpoints for this. /post-comments/ returns top-level comments on a public video, and /post-comment-replies/ expands the thread under any one of them. Both are plain HTTPS GET calls authenticated with an X-Api-Key header, both page with a cursor, and both return JSON directly - there is no job to submit and no dataset to fetch afterwards.

What a scraped comment contains

Each entry in the comments array carries the comment itself plus a nested user object identifying whoever wrote it:

A complete comment scrape

The comments endpoint keys off the public video URL, so you can go straight from a link someone pasted to a dataset:

curl "https://api.tikliveapi.com/post-comments/?url=https%3A%2F%2Fwww.tiktok.com%2F%40tiktok%2Fvideo%2F7460937381265411370" \
  -H "X-Api-Key: YOUR_API_KEY"

A trimmed response:

{
  "comments": [
    {
      "id": "7460951322251920150",
      "video_id": "7460937381265411370",
      "text": "please Mr. TikTok, don't sell it to Elon or Zucky 😭",
      "create_time": 1737138145,
      "digg_count": 878058,
      "reply_total": 93,
      "user": {
        "unique_id": "millennial.dominique",
        "nickname": "Dominique (Taylor's Version)",
        "region": "DE",
        "verified": false,
        "follower_count": 0
      }
    }
  ],
  "total": 353627,
  "cursor": 20,
  "hasMore": true
}

To walk the whole thread, keep calling with the returned cursor until hasMore is false. Add count to size each page: both endpoints accept up to 50 per request, and a page can come back with fewer. Then expand the comments with the highest reply_total:

curl "https://api.tikliveapi.com/post-comment-replies/?video_id=7460937381265411370&comment_id=7460951322251920150&count=50" \
  -H "X-Api-Key: YOUR_API_KEY"

Replies come back with the same fields as top-level comments except reply_total. The full pagination loop, including de-duplicating comments that repeat across adjacent pages, is worked through in fetching TikTok comments at scale with pagination.

What teams build with it

Scope and limits

Only comments on public videos are readable, exactly as a logged-out visitor would see them; comments on private accounts are not accessible. TikLiveAPI stores none of it: each request fetches the public comments from TikTok and forwards the response to you in real time. What arrives is the comment record described above plus the commenter's public identity - handle, nickname, avatar, region, language, bio (signature) and the linked-account fields TikTok includes (ins_id, twitter_id, twitter_name, youtube_channel_id, youtube_channel_title). A bio can hold whatever contact details its author chose to publish. Privacy duties attach to the copy you keep. If you store comment text and handles about identifiable people in the EU or UK, you need a lawful basis under GDPR Article 6, the CCPA can apply to data about California residents, and retention limits apply like any other personal data. Related reading: the comments API power-user guide and the broader TikTok scraper overview. Comments are one part of the posts category; the video API reference covers the rest, from single-video detail to feeds and search.