Instagram Comment Scraper with Sentiment
Pricing
from $1.75 / 1,000 results
Scrape Instagram comments from single post URLs or from an account's recent posts, with a sentiment label and score on every comment. $1.725 per 1,000 comments: 33% below the Free plan price of apify/instagram-comment-scraper, and the same flat rate on every plan.
Input URL
inputUrl
Optional
The line from directUrls that produced this row, exactly as it was written - a profile URL, a bare handle, or a single post URL. Present so a multi-target run can be split back into its targets. Profile rows, post rows and ERROR rows only: since 2026-09-10 a comment row is exactly apify/instagram-comment-scraper's twelve fields plus any the listing adds, and that reference row has no inputUrl - join a comment back to its target through postUrl. An error row always carries it, because an input that produced no data is exactly the row a caller needs to match back to the line they sent.
ID
id
Optional
Instagram's own identifier: the profile id on a profile row, the media id on a post row, the comment id on a comment row (never the commenter's account id, which is not collected; absent on snapshots collected before 2026-09-04). Always a string - these exceed what a JavaScript number holds exactly, and quoting them is what keeps the last digits.
URL
url
Optional
Canonical instagram.com URL of the profile or post this row describes. Null on an error row whose input never resolved to either - a string that is not an Instagram URL has no canonical form to report.
Published at
timestamp
Optional
ISO 8601 in UTC, ending in Z (2026-08-16T18:01:36.000Z). Absent when the feed carried no date - an unknown date is left out rather than rendered as 1970.
Likes
likesCount
Optional
-1 when the account hides its like counts, matching the convention of the other Instagram scrapers. Zero means zero.
Author
ownerUsername
Optional
On a post row, the account that published it. On a comment row, the account that wrote the comment.
Comment
text
Optional
The comment as written.
Post URL
postUrl
Optional
Permalink of the post this comment sits under.
Comment URL
commentUrl
Optional
Permalink of this comment, in the shape apify/instagram-comment-scraper writes: the post URL, /c/, then the comment id, with no trailing slash. Absent on snapshots collected before 2026-09-04, which carry no comment id to build it from.
Commenter's profile picture
ownerProfilePicUrl
Optional
Same value as owner/profile_pic_url, duplicated here exactly as apify/instagram-comment-scraper duplicates it. A SIGNED URL THAT EXPIRES in 4-5 days: it is a lease on the image, not a copy of it, so fetch it during the run or not at all. Collected since 2026-09-10 (COMPLIANCE.md section 3, the row of that date); null on any snapshot older than that, which means the record predates the field, not that the collection failed.
Replies
repliesCount
Optional
ALWAYS NULL, in this Actor and in apify/instagram-comment-scraper alike. Instagram renders no replies to a logged-out reader and does not count them either: child_comment_count was null on every node measured (217 nodes over three posts, 2026-09-04). A zero here would be an invented measurement.
Reply bodies
replies
Optional
ALWAYS NULL, for the reason repliesCount is: the logged-out comment surface serves top-level comments only. Present because the reference scraper writes the key on every row, so a consumer switching over keeps the column.
Parent comment
parentCommentUrl
Optional
ALWAYS NULL, for the reason replies and repliesCount are: every comment the logged-out surface serves is a top-level one, so none of them has a parent. Present because apify/instagram-comment-scraper writes the key on every row, and a consumer switching over keeps the column rather than having it disappear.
Comment author
owner
Optional
The reference scraper's nested author object, emitted with every one of its keys so a flattened export has the same owner/... columns in both Actors. owner/username, owner/id, owner/profile_pic_url and owner/is_verified carry values - the last three since 2026-09-10, and null on snapshots older than that. full_name, is_mentionable, is_private, profile_pic_id and latest_reel_media are null here and null in apify/instagram-comment-scraper's own output too, so a null there is parity, not a gap.
{ "username": "luv_yazzy", "profile_pic_url": "https://instagram.fmxp2-1.fna.fbcdn.net/v/t51.2885-19/518868862_1234.jpg", "is_verified": false, "id": "7058325994", "full_name": null, "is_mentionable": null, "is_private": null, "profile_pic_id": null, "latest_reel_media": null}Sentiment
sentimentLabel
Optional
How this comment reads. Absent when no classifier ran over the snapshot; null when it ran and had nothing to say.
Sentiment confidence
sentimentScore
Optional
0 to 1. A low score on a labelled comment means the classifier was unsure, not that the sentiment is weak.
Error code
error
Optional
Set only on an error row: the machine-readable slug for why this input produced no data, e.g. profile_not_found, post_not_found, collection_failed, input_refused. It is the same vocabulary the REST surface answers with, so one consumer can switch on both. A run writes one error row per input it could not collect, which is what makes the number of rows out equal the number of inputs in.
Error description
errorDescription
Optional
Set only on an error row: the sentence a human reads, saying what went wrong and usually what to do instead. Where error is for a program to branch on, this is for a person to act on.
Request error messages
requestErrorMessages
Optional
Set only on an error row: the raw messages collected while this input was being attempted, in order. A list even when there is one, so a consumer reading the first element never has to check the type first. Empty when the input was refused before anything was attempted.