Medium Scraper & API: Articles, Claps & Full Text Export

Medium Scraper & API: Articles, Claps & Full Text Export

Export Medium articles with their real numbers: claps, responses, reading time, word count, member-only flag, author followers and the full text as Markdown. Get articles by tag (the latest, or the most read of every month back to any date), by keyword search (up to ~1 000 per query), the whole history of an author or a publication, or by article link. No Medium account, no browser, no 10-article RSS limit.

{ "tags": ["python"], "tagSort": "top", "publishedAfter": "2026-07-01", "minClaps": 100 }

Filter by date range, minimum claps and free / member-only; articles you filter out are free. Schedule it with Only new articles to get every new article of a tag, an author or a keyword as an alert.

What you get

Real example, 25 October 2026: tag python, Most read, July to September 2026:

published	title	author	publication	claps	responses	read min	member-only	claps / 1k followers
2026-07-12	Eigenvalues for People Who Slept Through Linear Algebra	Kamrun Nahar	Towards AI	4,050	63	15	yes	1,202.9
2026-08-15	How The Fast Fourier Transform Actually Works	Kamrun Nahar	Towards AI	2,823	48	23	yes	838.4
2026-08-26	Structuring 40 Million Documents into an Agentic Knowledge Graph	Fareed Khan	Level Up Coding	1,753	22	58	yes	22.5
2026-07-15	Building GLM 5.2 744B Model in C to Run on 25GB RAM	Fareed Khan	Level Up Coding	1,408	8	72	yes	18.1
2026-09-21	Jev Clearly Explained	Fareed Khan	Level Up Coding	1,035	15	21	yes	13.3

Search "vector database", with the text:

title	author	published	claps	responses	read min	full text
Vector Databases Are Becoming a Feature, Not a Product	Tajamul Khan	2026-10-20	292	2	4	preview (member-only)
Vector Databases Explained: How AI Finds Relevant Information	Kehkasha	2026-10-07	159	1	12	3,135 words
Vector Databases: When Search Needs Meaning, Not Just Keywords	Aliakbar Hosseinzadeh	2026-10-19	73	1	5	1,169 words

Field	Description
id, title, subtitle, url	The article, its subtitle and link
authorName, authorUsername, authorUrl, authorFollowers, authorBio	The writer and their follower count
publicationName, publicationSlug, publicationUrl, publicationFollowers	The publication it appeared in (empty for personal posts)
publishedAt, updatedAt	First published and last updated (UTC)
claps, clappers, responses	Total claps, the number of people who clapped, the number of responses (comments)
clapsPer1kFollowers	Claps per 1,000 followers of the author: finds articles that did far better than the author's audience
readingTimeMinutes, wordCount	Reading time as Medium shows it ("12 min read") and the number of words
isMemberOnly	Behind the Medium paywall (member-only)
language, tags, imageUrl	Detected language, Medium tags, the cover image
markdown, text, isFullText	The article as Markdown (headings, links, lists, code blocks with language, images) and plain text. isFullText = false: a member-only article, you get the preview Medium shows to visitors
source, sourceType, rank	The tag, query, author or publication the row came from, and its place there
scrapedAt	Run time

How it works

1. Add one or more sources: Tags (artificial-intelligence, or a medium.com/tag/... link), Search queries, Authors (@name, a profile link or name.medium.com), Publications (better-programming, a medium.com/... link or a custom domain like javascript.plainenglish.io) or Article links
2. For tags, choose the order: Latest (newest first, as deep as the tag goes), Most read (the ~20 most read articles of every month in your date range; all time without dates) or Oldest
3. Optionally set Published after / before (2026-01-01 or 30 days, 6 months), Minimum claps and Member-only (all, free only, member-only only)
4. Set Max articles per source and run. Download as CSV, Excel, JSON or HTML. Three tabs: Articles, Engagement, Text

The article text is on by default; turn it off for faster runs when you need only the numbers.

Why this Actor

📚 No 10-article limit	Medium's RSS feeds give only the 10 latest articles. Here a tag goes back years, an author or a publication gives the whole history, a search up to ~1 000 articles
🏆 Most read of every month	The top articles of a tag for each month of any period: what really worked in your niche in 2024, 2025 or last quarter
👏 Real engagement	Claps, clappers, responses, reading time, word count and author followers on every row, plus claps per 1,000 followers to spot outliers
📝 Full text as Markdown	Clean Markdown of free articles, ready for RAG, LLM training sets, summaries or newsletters; member-only articles are flagged
🎯 Filters that cost nothing	Dates, minimum claps, free / member-only: filtered-out articles are not charged
🔔 New-article alerts	Only new articles returns just what appeared since the previous run of the same source
💰 $1.99 per 1,000 articles	Text included, no subscription

Get new Medium articles every day

1. Add your tags, authors, publications or keywords and turn on Only new articles
2. Save as a task and add a schedule, e.g. every morning
3. The first run returns the latest articles as a baseline. Every later run returns only articles not seen before: send them to Google Sheets, Slack or email with an integration

The memory is kept in your account (key-value store medium-scraper-state). Use a different Monitor name for each task.

Pricing

Pay per article in the results, no subscription needed. Apify Scale and Business plans pay less:

Event	Free and Starter plans	Scale plan	Business plan
Run start	$0.001	$0.001	$0.001
Article (one row, text included)	$0.00199	$0.00155	$0.00098

Task	Cost
20 latest articles of a tag	~$0.04
Most read Python articles of every month of 2025 (~230)	~$0.46
1,000 search results for one keyword, with text	~$2.00
Daily alert of new articles of 5 authors (~3 a day)	~$0.21 per month

Articles left out by your filters or by Only new articles are not charged.

Ready-made tasks

Open it, change the tag, and run:

Task	What you get
Top Medium Articles by Tag	The most read articles of a tag for every month of the last year, with claps, responses and text

FAQ

Where does the data come from? From medium.com itself, the same data any visitor sees, without logging in. Claps, responses and followers are Medium's own counts at the time of the run.

Do I get the text of member-only articles? No. Member-only (paywalled) articles give the preview Medium shows to visitors who are not members, and isFullText is false. All their numbers (claps, responses, reading time, words) are complete. Use Member-only = Free articles only when you need the full text of every row.

How many articles can I get? A tag: as many as it has with Latest (Medium tags have up to hundreds of thousands), about 20 per month with Most read. A search: up to ~1 000 per query (Medium's limit). An author or a publication: the whole history.

Why does "Most read" give ~20 articles per month? That is the list Medium keeps for each period. Set a longer date range for more months, or use Latest with Minimum claps to scan every article of the period.

What date formats work? 2026-01-31, 2026-01-31T10:00:00Z, or relative: 7 days, 2 weeks, 3 months, 1 year.

Do I need a Medium account or an API key? No.

Use with the API and AI agents

Run it via the Apify API from Python, Node.js or any HTTP client as your Medium articles API, or connect it to Claude, ChatGPT and other AI agents through the Apify MCP server: "Find the most clapped Medium articles about AI agents from the last 3 months and summarize what they have in common."

Other blog & content tools

Tool	What it does
Substack Scraper	Substack newsletters and posts with likes, comments and full text
YouTube Transcript Scraper	Transcripts of YouTube videos and channels as text
Audio & Video Transcriber	Podcasts and any audio or video file to text
Threads Scraper	Threads posts and profiles with likes and replies
Google Sheets Import & Export	Send the articles to a Google Sheet

Is it legal?

The Actor reads only public pages and data that Medium shows to any visitor, without logging in, and does not get around the paywall. Articles belong to their authors: respect copyright when you reuse the text, and follow data protection rules (such as GDPR) for the author names and bios in the results.