Threads scraper: profile posts, replies and search, without an account

What you get

One row per post, and a free row per profile.

Where Threads does not give a field, it is null, never guessed. The dataset has a Posts view, a Profiles view and an All fields view, and exports as JSON, CSV, Excel or XML, or through the API.

• The post: its ID, short code and URL, the text, the date posted, the language and topic when Threads states them, whether it is pinned, and whether it was edited.
• The author: handle, ID, display name and whether the account is verified.
• Engagement: likes, replies, reposts, quotes and reshares, as counts. No view count: Threads does not send one to a logged-out visitor.
• Replies and quotes: whether the post is a reply and to whom, and the URL of the post it quotes.
• Media and links: the media type, every image and video with its link and size, the accounts the text mentions, the links written in the text, and the link card. Media links are Threads' own signed links and expire: download what you need soon after the run.
• The profile row, free: handle, ID, name, bio, followers, verified, private or not, links and avatar.
• Where the row came from: the profile, post URL or keyword, its type, the position in that source and the time it was read.

Whether a media item is an image or a video is told from the renditions Threads sends, not from a type it states. Every run also writes a summary: what each source gave, where its list stopped and why, what was charged and what was free.

Threads profile scraper, replies scraper and search

Three ways in, any mix, one run.

• Posts per profile: 50 by default. Threads ends a profile's list at about 200 posts for a visitor without an account, whatever you ask for.
• Replies per post: 0 by default, which reads the post alone.
• Maximum posts in the run: an optional cap for the whole run, besides the spending limit you set on Apify.

{
  "profiles": ["natgeo", "https://www.threads.com/@nasa"],
  "maxPostsPerProfile": 200,
  "postUrls": ["https://www.threads.com/@nasa/post/DdXJKzbkqPg"],
  "maxRepliesPerPost": 100,
  "searchQueries": ["coffee"]
}

Measured in the cloud

400 posts from two profiles in 73 seconds.

The Actor sends plain HTTP requests, with no browser, and runs in 256 MB of memory. Runs of our own private build on Apify on 2026-11-10, logged out, through Apify's datacenter proxy:

One day and a handful of runs, not a track record. In the four runs of the current build, Threads refused none of 70 requests. Two things did go wrong that day, and the runs said so: on the earlier build, one search of three timed out and was lost, and since then a request that gets no answer is sent once more; and on a second profile, Threads answered an error after 222 posts, which ended that profile as partial with the reason (run hh3DEngH0uabFN8r6). Seven profiles of 50 posts took 128 seconds, and a handle that does not exist came back as a free row (run 9KXhWJasgPNsMPEkk): sources are read one after another.

Why this one

Six Threads scrapers on Apify, price by price.

Free plan prices, monthly users and the share of runs that succeeded over 30 days, as each Actor's public record showed them on 2026-11-10 (api.apify.com/v2/acts/futurizerush~meta-threads-scraper for the leader, the same address for each). "Success" is the share of runs that succeeded, not of posts delivered. We ran none of these Actors, and what each returns was not compared field by field or on the same input.

1,000 posts in one run cost $1.50 here and $2.52 at the leader on the Free plan; on many small runs its start fee weighs more. The leader's record shows a default memory of 4 GB; this Actor runs in 256 MB. Where we lose: the others have users, runs and a 30-day record, and ours is new. Two of the six are search Actors, and this one returns only about 20 posts per query; how deep theirs go was not measured by us. And this one has no view counts.