Instagram Scraper — profiles, posts and reels

Instagram hands a signed-out visitor its data through the endpoints its own site calls, and those endpoints move. The profile lookup currently errors out on any account with a business category set — most brands, most creators. So this reads whichever route is answering: the site's own lookup when it works, the rendered page and the account's post feed when it doesn't. A username gets you the profile row plus that account's recent posts, a post or reel URL gets you that single item. Everything comes back flat, one JSON row per thing, ready for a spreadsheet or a database.

No login, no cookie, no API key, no app review. Instagram serves profiles and public posts to a signed-out client and that's the route this uses.

What you get

Profile row (type: "profile"):

{
  "ok": true,
  "type": "profile",
  "username": "nasa",
  "id": "528817151",
  "fullName": "NASA",
  "biography": "Making the seemingly impossible, possible. ✨",
  "externalUrl": "https://www.nasa.gov/",
  "followersCount": 104400555,
  "followsCount": 92,
  "postsCount": 4887,
  "verified": true,
  "isBusinessAccount": true,
  "private": false,
  "profilePicUrl": "https://scontent-fra3-2.cdninstagram.com/...",
  "url": "https://www.instagram.com/nasa/"
}

Post and reel rows (type: "image", "video" or "carousel") carry: shortCode, url, caption, hashtags[], mentions[], likesCount, commentsCount, videoViewCount, displayUrl, videoUrl, images[], isVideo, dimensions, ownerUsername, ownerId, timestamp (ISO 8601) and accessibilityCaption.

hashtags and mentions are parsed out of the caption for you, so you don't have to regex them yourself downstream.

Input

{
  "usernames": ["nasa"],
  "resultsLimit": 50
}

Run it with no input at all and you get one labelled _sample row showing the output shape, charged nothing.

Limits

• Private accounts return the profile shell only. Follower counts, bio and private: true come through; the posts don't, because there's nothing public to read.
• Recent posts, not the full archive. Timeline paging stops at 60 pages, so very large accounts won't come back complete in one run. Use onlyPostsNewerThan to make the walk purposeful.
• postsCount comes back null on the fallback route. When the site's own profile lookup errors out, the rendered page it falls back to carries the follower and following counts but not the account's total post count. Everything else in the profile row is there.
• No comments, no follower lists, no stories. Those are separate scrapers.
• Media URLs expire. Instagram signs its CDN links with a short lifetime. Fetch the file while the run is going if you need to keep it; a URL you save today will 403 on you later.
• Items are deduplicated within a run, so nothing is returned or charged twice.

Billing

$0.00045 per item ($0.45 per 1,000), plus $0.001 when a run starts. Flat rate on every Apify plan. No volume tiers to climb, no minimum spend.

A profile row and a post row cost the same. You're charged only for rows that carry data — sample rows, diagnostic rows, date-filtered posts, duplicates and empty results are free. The start fee goes by run memory, one event per gigabyte, so it's $0.001 at the default 1 GB.