Instagram Comment Scraper

Instagram Comment Scraper pulls the comments under any public Instagram post or reel, with each comment's text, likes, reply count, time and flags such as pinned or liked by the account. Ask your AI assistant for the top comments on a post, or page through the whole thread from your code.

  • 20+ fields per comment
  • Tested
  • REST API + MCP
  • Billed per run

1Your request

Use the AI/ML API MCP Instagram Comment Scraper tool. Show the most liked comments on NASA's Voyager Golden Record post.

Post
instagram.com/p/DfkkknjD6vm (NASA)
Order
Popular
Page
First

2Result15 of 1,947 comments · 16 Nov 2026

#Comment and authorLikesRepliesMarked
1Somewhere in the universe, our story is still travelling. ✨️@ya••• · 10 Nov, 19:00 UTC6,93227Pinned
2Image descriptions: 1. The Golden Record…@nasa, the post owner · 10 Nov, 18:40 UTC4,77230Pinned
350 years for this craft to reach 1 light day. We really don’t fully understand how massive our universe is.@za••• · 10 Nov, 18:55 UTC7,50828
4Bro leaked our coords...@ch••• · 11 Nov1,1263
5(GIF, no text)@im••• · 15 Nov41GIF only
6When it’s released on Spotify?@aa••• · 10 Nov, 18:42 UTC6696NASA replied
7Proof that humanity’s legacy will always be art and curiosity. 🚀🎶@ni••• · 10 Nov, 18:46 UTC9408Liked by NASA
11I have the limited print vinyl from 10y ago. Impressive material@ko••• · 10 Nov, 18:56 UTC1,3524Liked by NASA
Source: Real result from 16 Nov 2026, 8 of the 15 comments on the first page shown. Commenter handles are masked on this page. Likes and replies are counts at the moment of the run.

What is Instagram Comment Scraper?

Instagram Comment Scraper reads the comment section of one public post and returns one record per comment. You give it a post link or shortcode and choose popular or recent order. Each call returns a page of comments, 15 in our test, plus a token for the next page, so you can stop after the top comments or walk the whole thread.

Every record carries the text, like count, reply count and time, along with flags for pinned, edited and liked by the post's owner. The first page also returns the post caption and the total number of comments, so you know the size of the thread before you collect it.

You ask

Show the most liked comments on NASA's Voyager Golden Record post.

It runs

Instagram Comment Scraper

From your AI assistant over MCP or from your code over REST

You get

  • Comment
  • Engagement
  • Commenter
  • Post and paging
  • Mine questions from launch postsPull the comments on a product reveal and list what people ask about price, sizes and release dates.
  • Measure the reaction to a campaignCompare comment volume, likes and replies across every post of one campaign.
  • See what the brand answeredFind the comments an account liked or replied to, and the popular questions it left open.
  • Collect customer languageSave the exact words people use about a product for ad copy, an FAQ or a landing page.
  • Let your agent read the threadHave Claude or ChatGPT group the comments by topic and quote examples in the same chat.

What data does the Instagram comment scraper return?

Each comment comes back as one record, and each page carries a few fields about the post itself. These are the fields our test call returned.

  • Comment

    1. Comment ID
    2. Text
    3. Time posted
    4. Position in the list
    5. Edited flag
    6. Hashtags and mentions in the text
    7. GIF attachment with title and link
  • Engagement

    1. Like count
    2. Reply count
    3. Preview of the first reply
    4. Liked by the post owner
    5. Pinned flag
  • Commenter

    1. Username and display name
    2. User ID
    3. Private or public account
    4. Verified badge
    5. Profile picture
    6. AI-generated profile label
  • Post and paging

    1. Post caption with its hashtags
    2. Total comment count
    3. Number of pinned comments
    4. Token for the next page

Fields appear when Instagram has them: a GIF reply has no text, and only some comments carry a reply preview. For every reply under one comment, use Instagram Comment Replies. Commenters are private people, so keep their handles only when you have a reason to.

Sample output from a NASA post

The table at the top of this page is a real call on 16 November 2026: popular comments on NASA's 10 November carousel about the Voyager 1 Golden Record. The first page returned 15 comments, and the post reported 1,947 in total. Below the chart is one comment in full, with the reply NASA posted under it.

Likes in this run, highest first
  1. 50 years for this craft to reach 1 light day. We really don’t fully understand how massive our universe is.7.5K
  2. Somewhere in the universe, our story is still travelling. ✨️6.9K
  3. Image descriptions: 1. The Golden Record…4.8K
  4. I have the limited print vinyl from 10y ago. Impressive material1.4K
  5. Bro leaked our coords...1.1K
  6. Proof that humanity’s legacy will always be art and curiosity. 🚀🎶940
  7. When it’s released on Spotify?669
  8. (GIF, no text)4

One result in full

Comment 6 in full

Text
When it’s released on Spotify?
Commenter
@aa••• (private account, not verified)
Posted
10 Nov 2026, 18:42 UTC, two minutes after the post
Likes
669
Replies
6
First reply
From @nasa, 843 likes: “@aa••• You can explore the music and sounds on the Golden Record here: go.nasa.gov/GoldenRecordSounds”
Pinned or edited
No
Position
6th in popular order
Next page
Token returned for the next 15 comments
Commenter handles are masked on this page. The tool returns each commenter’s username, display name, profile picture and account flags.

What one page of comments tells you

  • The total on the page, 1,947, matches the comment count on the post. At 15 comments per page, the full thread takes about 130 calls, so decide how deep you need to go.

  • The top three comments hold 19,212 of the 25,838 likes on this page, 74%. Popular order puts the comments people agreed with first, which is usually what a summary needs.

  • 10of 15

    10 of 15 comments were written within 36 minutes of the post going live. NASA's own pinned comment, with image descriptions, came within a minute, so drop owner comments when you measure audience reaction.

  • 2of 15

    2 of 15 comments are GIFs with no text, and NASA liked 4 of the 15. Filter empty text before you count themes.

Read an Instagram comment section in four steps

  1. Connect AI/ML API

    Click Try it now and add the MCP server to Claude, ChatGPT, Cursor or another client. You sign in once.

  2. Paste the post link

    Copy the link of a public post or reel and say whether you want the top comments or the newest, and how many.

  3. Get the comment table

    Your assistant returns each comment with likes, replies and time, with pinned and owner-liked comments marked.

  4. Group, quote and export

    Ask for themes, unanswered questions or a CSV, or have it fetch the next pages before it summarizes.

Prompts to copy

  • Use the AI/ML API MCP Instagram Comment Scraper tool. Get the most popular comments on instagram.com/p/DfkkknjD6vm and show text, likes and replies in a table.

  • Use the AI/ML API MCP Instagram Comment Scraper tool. Collect the newest 60 comments on this reel and list every question about price or shipping: [post link]

  • Use the AI/ML API MCP Instagram Comment Scraper tool. Find the latest post on instagram.com/nike, then group its top 45 comments into praise, complaints and questions with two quotes each.

  • Use the AI/ML API MCP Instagram Comment Scraper tool. Compare the top comments on these two launch posts and tell me which product drew more questions: [link 1] [link 2]

Settings for a comment scrape

The scraper has three settings. Your assistant fills them in from your request and handles the paging when you ask for a number of comments.

SettingWhat it controlsTip
Post link or shortcodeThe public post or reel to read, as a full URL or its short codeThe code is the part after /p/ or /reel/ in the link
OrderPopular puts the most liked comments first, recent puts the newest firstPopular for summaries, recent for catching new comments
Next-page tokenReturned with every page and passed back to get the following oneStop when you have the count you need

Pinned comments come first. On the NASA post, two of the 15 were pinned and one was written by NASA itself, so treat owner comments separately from the audience.

Ways teams use Instagram comments

  • Launch feedback

    Read the top comments on a product reveal and list the questions about price, sizes and release dates for your support team.

    TextLikesTime

  • Ad and landing page copy

    Collect the phrases people use about your product and reuse their words in headlines and objections.

    TextLikes

  • Community management review

    See which comments your brand liked or answered, and which popular questions got no reply at all.

    Liked by ownerRepliesPinned

  • Creator due diligence

    Check whether a creator's comment section holds real conversation or mostly emoji and GIF replies before a paid deal.

    TextGIFLikes

  • Competitor reaction

    Compare comment themes on a competitor's launch with yours, using the same order and the same number of pages.

    TextLikesReplies

  • Giveaway checks

    Export the comments on a contest post with time and handle to verify entries and the closing time.

    UsernameTimeText

Instagram comment scraper API

In your AI assistant

MCP · sign in once with your AI/ML API account

MCP setup guide ↗

In your code

Backend, script or scheduled job

REST · your AI/ML API key, JSON records back

Over REST, the scraper takes a post link, an order and an optional page token, and returns JSON with the comment list and the token for the next page. Loop until you have the number of comments you need.

Use MCP when an analyst reads comments in a chat and REST when a job checks new comments on your posts every hour. Both run on one AI/ML API balance and need no Meta app or Instagram login. Start with the Toolbox REST quickstart.

Other ways to collect Instagram comments

Comments are easy to read and hard to collect at volume. Here is how the usual methods compare.

Other ways to collect Instagram comments: Instagram Comment Scraper on AI/ML API, Copy comments by hand, Instagram API from Meta, Your own scraper
AI/ML APICopy comments by handInstagram API from MetaYour own scraper
SetupConnect once with your AI/ML API accountNoneMeta developer app, business account, app reviewLogged-in sessions, paging logic and steady upkeep
Inside an AI chatYes. Over MCPNo. NoOnly if you build itOnly if you build it
Best forAny public post, from chat or codeA few quotes for one slideComments on your own postsTeams with engineers on call

Pricing

Pay as you goBilled per run from your AI/ML API balance

  • No subscription
  • No seat fee
  • One balance for every tool

Each call returns one page of comments and is billed per run from your AI/ML API balance, with no subscription. Our 15-comment test call cost less than one cent. Long threads take many pages, so check the price in your dashboard before you export thousands of comments.

See pricing

FAQ

Is it legal to scrape Instagram comments?

The tool reads comments on public posts without logging in. Commenters are private people, so privacy laws such as GDPR apply when you store their handles or quote them, and Instagram's terms apply to the service. Ask a lawyer about your specific case.

How many comments can I get from one post?

As many as the post has, page by page. Our first page held 15 comments and a token for the next one, so a post with 1,947 comments takes about 130 calls.

Does it include replies to comments?

Each comment carries its reply count and sometimes a preview of the first reply. For the full reply thread under one comment, use Instagram Comment Replies.

Where do I get the post link?

Open the post and copy its link, or ask your assistant to list an account's recent posts with Instagram Scraper and pass the one you want.

How is this different from the Instagram API from Meta?

Meta's API returns comments on media from accounts you manage and needs a developer app with review. Instagram Comment Scraper reads comments on any public post, runs on your AI/ML API balance and works inside AI assistants.

Can my AI assistant analyze the comments?

Yes. Add the AI/ML API MCP server to Claude, ChatGPT, Cursor or another client, ask for the comments, then ask for themes, questions or quotes in the same chat.

Does it return sentiment scores?

No. It returns the comments as written, with likes and replies. Your assistant can sort them by tone, and you can check its groups against the quoted text.

Can I export comments to CSV?

Ask your assistant to return the table as CSV, or write the JSON from the REST response to a file in your code.

How current are likes and replies?

Each call reads the post at that moment. Comments keep collecting likes for days, so a later run can change the popular order.

Read your first comment section

Paste a post link and get the comments as a table in your chat.