Vinted Scraper: Every Listing, Past the 960 Cap avatar

Vinted Scraper: Every Listing, Past the 960 Cap

Pricing

from $0.50 / 1,000 listings

Go to Apify Store
Vinted Scraper: Every Listing, Past the 960 Cap

Vinted Scraper: Every Listing, Past the 960 Cap

Scrape every Vinted listing past the 960-per-search cap: 11,150 relevant listings on a search the site stops at 960. Keywords or URLs on 7 sites (FR, DE, ES, IT, PL, UK, US), total price with fees, only listings that contain your words, photos, seller, full description, new-listings monitoring.

Pricing

from $0.50 / 1,000 listings

Rating

0.0

(0)

Developer

Alexis Morain

Alexis Morain

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

0

Monthly active users

4 days ago

Last modified

Features

Limited permissions

Share

What does Vinted Scraper do?

Vinted Scraper exports every listing of a Vinted search, not only the first 960 that the site shows. Give it keywords and countries, or paste catalog URLs, and you get one row per listing: title, brand, size, condition, price, the Buyer Protection fee and the total price, favourites, photos, seller id and business flag. If you ask for it, each row also carries the full description, category, colour, posting time and the seller's feedback.

It works as a Vinted API for the public catalog: it reads the same pages the Vinted website shows to a visitor, over plain HTTP. No browser, no login, no cookies from your account.

It covers the 7 Vinted sites: vinted.fr, vinted.de, vinted.es, vinted.it, vinted.pl, vinted.co.uk and vinted.com. Each row keeps the currency of its site (EUR, PLN, GBP, USD).

Whatever the search, Vinted returns at most 960 listings: ten pages of 96, and the eleventh page is empty. A search for nike on vinted.fr shows 960 listings while the category counts of the search page add up to more than 300,000. A scraper that reads the ten pages loses the rest without telling you.

This actor notices when a search hits the cap and cuts it into smaller searches until every piece is under 960. It cuts in this order: first by category, then by price range, then by condition, and last by colour. It reads the pieces in parallel, removes the listings that appear in two pieces, and reports at the end how many listings Vinted announced and how many it delivered.

Measured on 7 and 8 November 2026, on the Apify platform (vinted.fr):

SearchVinted announcedOne search (what the site shows)Read by the actorFiltered out by Only listings that contain every search wordDelivered
tamagotchi14,16096013,9972,84711,150
ipod classic1,2359601,228620608
salomon xt-6 (before that option existed)2,7209602,707none2,707
new balance 530 (before that option existed)18,41996018,166none18,166

"Announced" is the total that the category counts of the search page add up to when no filter is set. It is Vinted's own count, and Vinted's search is fuzzy: it matches listings that only partly contain your words. That is why the actor reads almost all of the announced listings and then, by default, delivers fewer: it keeps only the listings whose title or brand contains every word of your search (see below). In these runs no cell stayed at 960 after the last split and no cell failed. The gap between announced and read (0.5% to 1.4%) is explained in the FAQ.

Why use Vinted Scraper?

  • Resellers and sourcing. Watch a brand, a model or a category across countries for underpriced listings, with the total price including the Buyer Protection fee. Run it in New items only mode on a schedule and pay only for what is new.
  • Price and market research. Count listings and read asking prices by brand, model, condition or country. A search is no longer capped at the 960 most recent ones, so a median price rests on the whole market, not on a slice.
  • Resale and arbitrage between countries. The same model on vinted.fr, vinted.de and vinted.pl in one run, each row with its currency.
  • Brand and trend monitoring. Track how many listings of a brand appear each day and how favourites evolve.
  • Pipelines. Rows go to the dataset in JSON, CSV or Excel, to Google Sheets, Slack, Zapier, Make or n8n, and a webhook tells you when a run ends. Apify adds scheduling, an API for every run, proxy rotation and monitoring.

What data can Vinted Scraper extract?

FieldExampleDescription
id10711433187Vinted listing id, unique per site. Use it to deduplicate across runs
urlhttps://www.vinted.fr/items/10711433187-cargo-nike-noirLink to the listing
title, brandCargo Nike noir, NikeTitle and brand shown on the card
size, conditionXS, Neuf sans étiquetteSize (null when the listing has none) and condition, in the language of the site
price40Price asked by the seller, in the currency of the site
serviceFee2.7Buyer Protection fee added on top
totalPrice42.7Price plus fee: what the buyer pays for the item, shipping not included
currencyEURISO currency of the site
favouriteCount0Members who favourited the listing
isPromotedfalsetrue when the seller paid to promote the listing
sellerId, sellerIsBusiness312278158, falseSeller's Vinted member id and professional flag
photos, thumbnailUrl, dominantColor["https://images1.vinted.net/..."], #C7C5C4Full-size photo URLs, small cover photo and the dominant colour of the cover
domain, countrywww.vinted.fr, FRSite of the listing and its country (GB for vinted.co.uk, US for vinted.com)
searchQuery, searchUrlnike, https://www.vinted.fr/catalog?search_text=nike&order=newest_firstThe search that found the listing
scrapedAt2026-11-08T02:09:45ZWhen the row was collected

With Add the item page (details), each row also has:

FieldExampleDescription
descriptionCargo Nike noir porter une seule fois...The full text written by the seller
brandId, categoryId53, 583Vinted ids, usable in the Brand ids and Category ids inputs
categoryHommes PantalonsCategory path in the language of the site
color, materialNoir, nullDeclared by the seller. material is empty on many listings
uploadedTextIl y a une minuteWhen the listing was posted, as Vinted writes it ("a minute ago"), in the language of the site. It is not a timestamp
sellerLoginxcii.tcheumiiSeller's login
sellerFeedbackCount, sellerFeedbackReputation12, 1Number of reviews of the seller and the reputation value Vinted reports
canBuy, isReserved, isHiddentrue, false, falseState of the listing
detailsFetchedtruefalse on a row whose page was gone or unreachable

The list rows do not carry the posting date: Vinted does not give it in the search results. To get it, even as relative text, use the details option.

How to scrape Vinted

  1. Click Try for free.
  2. Type one or more Keywords (for example nike or levi's 501) and pick the Countries. Or paste Vinted catalog URLs into Vinted catalog URLs: the keyword, price range, category, brand, condition and colour filters are read from the URL.
  3. Set Max items per search. It is also your cost cap: each item is one charged result.
  4. Leave Go beyond 960 items on.
  5. Press Start, then download the dataset as JSON, CSV or Excel.

To watch a search, tick New items only and add a schedule.

How much will it cost to scrape Vinted?

The actor is billed per result: a start fee of $0.002 per GB of memory (512 MB is the default and counts as one event), plus one charge per listing. A listing read with details costs more than a plain one. Prices fall with your Apify plan:

PlanListing, per 1,000Listing with details, per 1,000
Free$1.00$2.00
Bronze$0.90$1.80
Silver$0.80$1.60
Gold$0.50$1.00

Examples at the Free plan price: the 11,150 listings of tamagotchi cost about $11.15 plus the start fee. 500 listings with details cost $1.00. In New items only mode, a listing delivered once is never charged again.

The Apify platform usage of a run is small and is included: the 11,150-listing tamagotchi run used $0.074 of platform usage in total, which is about $0.007 per 1,000 listings delivered. A run with details costs more in usage because it opens every listing page: 100 listings with details used $0.008. Max items per search and the spending limit you set on a run stop it at the cap you choose. The default of 1,000 items per search limits a first test to $1.00 at the Free price. Listings dropped by the keyword filter are not charged.

Input

See the Input tab for every option. The ones worth knowing:

FieldWhat it does
searchUrlsVinted catalog pages, one per line, from any of the 7 sites. Keyword, price range, category, brand, condition and colour are read from the URL. Size and material filters are not supported and are ignored (the log says so). Item, member and brand pages are rejected with a message
searchQueriesOne search per keyword and country
countriesfr, de, es, it, pl, co.uk, com. Empty with keywords means France
priceMin, priceMaxSeller's price range in the currency of each site, fees not included. Apply to the keyword searches
catalogIds, brandIdsVinted category and brand ids. The id is in the catalog URL after catalog[]= or /catalog/, and after brand_ids[]=. Several ids mean any of them
conditionsNew with tags, New without tags, Very good, Good, Satisfactory. Empty means all
maxItemsItems per search. 0 means no limit. Default 1,000
strictMatchOn by default. Delivers only listings whose title or brand contains every word of the keywords. Dropped listings are never charged
splitBeyondLimitOn by default. Off reads the first 960 only, in fewer requests
includeDetailsOpens every listing page. One extra request per item
newOnly, registryKeyReturns only listings not delivered by a previous run of the same searches
webhookUrlJSON summary posted when the run ends
maxConcurrencyParallel connections, 1 to 20, default 6
proxyConfigurationApify datacenter proxy by default

Filters set as inputs (priceMin, catalogIds, conditions and the others) apply to the keyword searches. A pasted URL keeps its own filters.

Example inputs

A keyword on three sites, with a price range:

{
"searchQueries": ["stone island"],
"countries": ["fr", "de", "it"],
"priceMin": 20,
"priceMax": 300,
"maxItems": 5000
}

Pasted catalog URLs (a category on vinted.fr, a keyword on vinted.co.uk):

{
"searchUrls": [
"https://www.vinted.fr/catalog?search_text=jordan&catalog[]=1231",
"https://www.vinted.co.uk/catalog?search_text=barbour"
],
"maxItems": 0
}

One brand across a category, only items in good condition, with the full listing:

{
"searchQueries": ["veste"],
"countries": ["fr"],
"brandIds": ["53"],
"conditions": ["New with tags", "Very good"],
"includeDetails": true,
"maxItems": 500
}

A daily watch, with a webhook:

{
"searchQueries": ["new balance 530"],
"countries": ["fr"],
"newOnly": true,
"maxItems": 0,
"webhookUrl": "https://hooks.example.com/vinted"
}

Output

You can download the dataset in various formats such as JSON, HTML, CSV, or Excel. A listing read with details (a real row from a run of 8 November 2026):

{
"id": 10711433187,
"url": "https://www.vinted.fr/items/10711433187-cargo-nike-noir",
"title": "Cargo Nike noir",
"brand": "Nike",
"size": "XS",
"condition": "Neuf sans étiquette",
"price": 40,
"serviceFee": 2.7,
"totalPrice": 42.7,
"currency": "EUR",
"favouriteCount": 0,
"isPromoted": false,
"sellerId": 312278158,
"sellerIsBusiness": false,
"photos": [
"https://images1.vinted.net/t/05_019b1_bWqAujuTMqja9ztpddSQPPiU/f800/67f7fe9b.webp?s=80cdba2c25c48d976984f1dc8804d2280ff1194f"
],
"thumbnailUrl": "https://images1.vinted.net/t/05_019b1_bWqAujuTMqja9ztpddSQPPiU/310x430/67f7fe9b.webp?s=6c46683d5ae592481407cd1dead462f01ce371d7",
"dominantColor": "#C7C5C4",
"domain": "www.vinted.fr",
"country": "FR",
"searchQuery": "nike",
"searchUrl": "https://www.vinted.fr/catalog?search_text=nike&order=newest_first",
"scrapedAt": "2026-11-08T02:09:45Z",
"description": "Cargo Nike noir porter une seule fois taille xs mais taille a peu près comme un S ( prix négociable )",
"brandId": 53,
"categoryId": 583,
"category": "Hommes Pantalons",
"color": "Noir",
"material": null,
"uploadedText": "Il y a une minute",
"sellerLogin": "xcii.tcheumii",
"sellerFeedbackCount": 12,
"sellerFeedbackReputation": 1,
"canBuy": true,
"isReserved": false,
"isHidden": false,
"detailsFetched": true
}

Without details the same row has description, brandId, categoryId, category, color, material, uploadedText, sellerLogin, sellerFeedbackCount, sellerFeedbackReputation, canBuy, isReserved and isHidden set to null, and detailsFetched set to false.

The run report

Every run saves a record named OUTPUT in its key-value store (the Run report link of the Output tab). For each search it gives the announced total (Vinted's count), the number of cells, the requests, the listings read (retrieved), the listings dropped by the keyword filter (filteredOut), the listings delivered, the parts skipped as off-topic (offTopicCells), the cells that stayed at 960 after every split and the cells that failed. The top level also carries delivered and filteredOut for the whole run. At the top, complete says whether the whole read went through. The status message of the run shows the same figures on one line, with the network totals: requests, rate-limited requests, IP rotations, megabytes.

Run it from your code

Python:

from apify_client import ApifyClient
client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("alexmorain/vinted-scraper").call(run_input={
"searchQueries": ["new balance 530"],
"countries": ["fr", "de"],
"maxItems": 5000,
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
print(item["totalPrice"], item["currency"], item["title"], item["url"])

JavaScript:

import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: '<YOUR_APIFY_TOKEN>' });
const run = await client.actor('alexmorain/vinted-scraper').call({
searchUrls: ['https://www.vinted.fr/catalog?search_text=salomon%20xt-6'],
includeDetails: true,
maxItems: 300,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items.length);

cURL:

curl -X POST "https://api.apify.com/v2/acts/alexmorain~vinted-scraper/run-sync-get-dataset-items?token=<YOUR_APIFY_TOKEN>" \
-H "Content-Type: application/json" \
-d '{"searchQueries": ["tamagotchi"], "countries": ["fr"], "maxItems": 100}'

The synchronous endpoint waits for the run to end, so use it for small runs. For thousands of listings, start the run and read the dataset when it finishes.

Only listings that contain every search word

Vinted's search is fuzzy. When a search is rare or narrowed by a price range, Vinted pads the results with unrelated listings: a search for polly pocket between 300 and 310 EUR returns a games console. On ipod classic, only 56% of the first 960 listings have both words in the title or brand. You would pay for the rest.

So Only listings that contain every search word (strictMatch) is on by default. A listing is delivered only if its title or brand contains every word of your keywords. Case, accents and simple plurals are ignored, a word matches the start of a word (classic matches classique), and a number such as 530 also matches inside MR530SG. Dropped listings are never delivered or charged, and the run report counts them as filteredOut. When the first page of a part of the search has almost no listing that matches (2% or less), the actor does not read that part further and names it in offTopicCells.

What it can lose: listings whose title does not carry your words, for example an iPod 6th generation 80 GB when you searched ipod classic. On ipod classic we estimate those at about 2.5% of the listings read (27 of the 960 first rows). On tamagotchi we found none in a sample of 22 dropped rows. These are estimates from small samples, not counts. If you want exactly what Vinted returns, turn the option off. It applies to keywords, and to the keyword of a pasted URL. A search made of filters only (a category or a brand without keyword) is not filtered.

Tips

  • Set maxItems first. A broad keyword such as nike holds hundreds of thousands of listings. Start with 1,000, look at the rows, then raise it. maxItems counts per search, so three countries with maxItems 1,000 can deliver 3,000 rows.
  • Narrow the search rather than raising the cap. A category id, a brand id or a price range cuts the number of requests and the cost for the same listings you care about.
  • Use the proxy default. The actor uses the Apify datacenter proxy and changes IP whenever Vinted refuses one. Vinted answers about 30 requests a minute per IP, so the speed comes from parallel connections, each on its own IP. About 5% of the requests meet an HTTP 403 on the way and are replayed on a fresh IP. Residential proxy also works but costs far more: we measured about $0.14 per 1,000 listings in proxy traffic, against about $0.006 per 1,000 for the whole run on datacenter.
  • Mind the memory on huge reads. At the default 512 MB a single run held about 100,000 delivered listings before the container ran out of memory (memory grows by about 3 KB per listing read). Above 80,000 listings in one run, set the memory to 1,024 MB or more in the run options (the start fee is charged per GB), or split the search with price or category filters. We have not run past 100,000 listings in one run.
  • Raise maxConcurrency for big reads. The measured speed on 16 connections is about 3,600 listings per minute. The default of 6 connections is slower, and 20 is the maximum.
  • Details cost one request each. Add them only when you need the description or the seller's feedback. Do a first run without details to select what you need, then read the details of those.
  • Read the status message. A run that stopped early, a search that failed or a cell that stayed at 960 is written there and in the run report.

Monitoring with New items only

The first run reads everything and then records that the read was complete. The following runs read the newest pages of each search, one by one, and stop at the first page where every listing was already delivered. If more than 960 listings are new since the last run, the actor goes back to the full read for that search. Listings already delivered are never charged twice.

A run that stops early (maxItems reached, spending limit, a search that failed) does not record a complete read, so the next run reads everything again. For a watch, leave maxItems at 0. The memory is a named key-value store, vinted-registry, with one record per set of searches. Change a search (another keyword, another price range) and the actor starts a new memory. Use registryKey to give a name to a watch, to keep several independent watches on the same search, or to change the search without losing the history.

On a test search (tamagotchi connexion, vinted.fr), the first run delivered 127 listings. The second run, a minute later, read one page, found 96 listings it already knew and delivered none, at no charge.

Webhook

If you set webhookUrl, the actor posts a JSON summary when the run ends: actor, runId, datasetId, mode (all or new-only), status (succeeded, partial or failed), searches, failedSearches, articles (the listings delivered), skippedAlreadySeen and chargeLimitReached. The URL is stored as a secret and never logged.

FAQ

How do I get more than 960 Vinted results?

You split the search until each piece is under 960, then merge. Vinted Scraper does it for you. When the first page of a search says 960, it reads the category facets of that page and opens one search per sub-category, and does the same inside each sub-category until it reaches the leaves. A piece that is still at 960 is cut by price (at the median price of its first page), then by condition, then by colour. The pieces are read in parallel and the duplicates are removed by listing id.

The actor also checks that the sub-categories cover their parent. Some listings sit at the parent level only and belong to no child. When too many listings of the parent's first page are missing from its children, the parent is read again, cut by price.

Why is there a gap of 1 to 1.5% between announced and delivered?

The four measurements above read 98.6% to 99.5% of the announced total (13,997 of 14,160, 1,228 of 1,235, 2,707 of 2,720, 18,166 of 18,419). The search is a live feed sorted by newest first, and listings are posted, sold and removed while the run goes. On nike on vinted.fr, 34 of the 95 listings of the first page had changed 25 seconds later. When new listings arrive on top, every page pushes its last listings onto the next page (these are the duplicates the actor removes), and a listing sold during the read can slide between two pages and be missed. The actor reads one more page when a piece has fewer distinct listings than announced, which recovers the listings pushed past the last page. We did not measure which part of the remaining gap comes from removals. The announced figure is also taken from the facets at the start of the read and is not updated.

How fast is it?

On Apify datacenter proxy with 16 connections, we measured about 3,600 listings per minute: 45,000 listings of nike on vinted.fr in about 12 minutes, in 831 requests. A full read of 18,166 listings took 8.8 minutes, and tamagotchi (13,997 read) 7.2 minutes. With the default of 6 connections it is slower. Without a proxy the run has one IP and reads about 30 requests a minute. With details, add one request per listing, spread over the same connections.

Does it work on all Vinted countries?

On the 7 sites listed above. Each one was tested in one run: 50 listings per site, with the right currency (EUR on vinted.fr, vinted.de, vinted.es and vinted.it, PLN on vinted.pl, GBP on vinted.co.uk, USD on vinted.com). Other Vinted sites, such as the Lithuanian or Dutch one, are not supported.

Can I search without a keyword?

Yes, with a category id or a brand id: catalogIds or brandIds with countries, or a catalog URL such as https://www.vinted.fr/catalog/1231-sneakers. A search with neither a keyword nor a category nor a brand is refused with a message, because it would be the whole site.

Can I filter by size?

Not yet. Size and material filters in a pasted URL are ignored and the log says so. Keep the rows and filter on size after the run.

What happens if my run is migrated to another server?

The Apify platform sometimes moves a run to another server, which restarts the actor with the same input and the same storages. The actor keeps a record of the listings it already delivered and of how many each search delivered. After a restart it skips them (they are neither pushed again nor charged again) and maxItems counts them: a restarted run does not go over the cap. It does read the searches again from the start, so a migration can lengthen a big run.

What does a failed item page do?

With details on, a listing whose page is gone (HTTP 404) or unreachable is delivered with detailsFetched set to false and charged as a plain listing. In a test of 100 listings with details, 99 pages were read.

Why does isPromoted matter?

Promoted listings are paid placements and can appear out of order in the feed. In New items only mode the actor never uses them to decide that a page is already known.

The actor reads public catalog pages that anyone can open without an account. Vinted's terms of use restrict automated access to its site, so check them against your use. Our Actors are ethical and do not extract any private user data, such as email addresses, gender, or location. They only extract what the user has chosen to share publicly. We therefore believe that our Actors, when used for ethical purposes by Apify users, are safe. However, you should be aware that your results could contain personal data. Personal data is protected by the GDPR in the European Union and by other regulations around the world. You should not scrape personal data unless you have a legitimate reason to do so. If you're unsure whether your reason is legitimate, consult your lawyers.

The rows carry the seller's member id, and with details the seller's login. Treat them as personal data.

What if Vinted changes?

The actor reads the data embedded in Vinted's catalog pages, which can change without notice. If a run starts failing or returning less, open the Issues tab and report it with the search you ran.

Support

Questions, a field you miss, a search that comes back short: use the Issues tab of this Actor. The API tab has the endpoints to start and read runs from your own code.

Guide and use cases

Use cases for outbound and automation, the output fields, pricing and common questions are on the Vinted Scraper page of my site. The Classifieds and real estate family page puts it side by side with the other scrapers of the same family.