Vinted Deep Scraper — Beyond the 960-Item Cap

Unofficial tool. Not affiliated with, endorsed by, or sponsored by Vinted.

Extract Vinted listings — price, brand, size, condition, favourites, photos and seller — from any of 24 Vinted marketplaces, as clean JSON, CSV or Excel.

The problem this solves

Vinted's catalog will not give you more than 960 items for a single query, no matter how many results actually exist. The page size is capped at 96, page 11 returns an error, and the reported total freezes at 960 — so the API stops telling you the truth well before you have the data.

Most Vinted scrapers inherit that ceiling and quietly stop there.

This Actor splits your query into disjoint price slices, each of which gets its own 960-item budget, and keeps subdividing any slice that is still saturated. Duplicates across slice boundaries are removed before anything is saved.

Measured on the query nike (vinted.fr, 11 August 2026):

Mode	Unique items returned
Standard (what the API gives you)	791
Deep scan	3 682

That is 4.6× more data from the same query, and the run had not exhausted the price range. Yield varies with how deep your query's inventory actually goes.

What you get per listing

{
  "id": "9634976009",
  "title": "T-shirt Nike",
  "url": "https://www.vinted.fr/items/9634976009-t-shirt-nike",
  "brand": "Nike",
  "size": "L",
  "condition": "Très bon état",
  "price": 5.0,
  "buyerProtectionFee": 0.95,
  "totalPrice": 5.95,
  "currency": "EUR",
  "favouriteCount": 2,
  "viewCount": 0,
  "isPromoted": false,
  "photoUrl": "https://images1.vinted.net/t/06_01e82/f800/1786455915.jpeg",
  "photoCount": 3,
  "sellerId": "285557605",
  "sellerLogin": "matheoh7",
  "sellerUrl": "https://www.vinted.fr/member/285557605-matheoh7",
  "sellerIsBusiness": false,
  "marketplace": "www.vinted.fr",
  "matchedQuery": "nike",
  "priceBucket": "3-5"
}

price is what the seller asks. totalPrice is what the buyer actually pays, buyer protection included — the number you need for any resale or arbitrage calculation, and the one that is missing from a plain listing export.

Input

Field	What it does
Search queries	One or more keywords. Leave empty to browse a filtered catalog with no keyword.
Marketplace	Which of the 24 country sites to query. Prices come back in that market's currency.
Maximum items	Hard stop for the run. You are charged only for items actually saved.
Deep scan	Breaks the 960-item cap by price splitting. On by default.
Sort by	Relevance, newest first, or price ascending/descending.
Minimum / maximum price	Restrict the range. Deep scan splits inside whatever range you set.
Category / brand / size / condition / colour IDs	Copy the catalog_ids, brand_ids, size_ids, status_ids or color_ids values straight out of a filtered Vinted search URL.

Cost

Priced per item saved. Nothing is charged for items that are not delivered: the run stops collecting as soon as your spending limit is reached, and the counter in the log always matches the number of rows in the dataset.

Notes and limits, stated honestly

• Deep scan is switched off automatically when maxItems is 960 or less. Below the cap it would add nothing and would bias your sample: slices are walked cheapest-first, so a small budget would return only the low end of the market instead of the most relevant results.
• Deep scan costs more requests than a flat run — roughly one request per 55 items delivered in testing — because saturated slices have to be probed before they are harvested.
• Sold items are not returned. Vinted's catalog only exposes active listings.
• No login, no proxy, no browser. The Actor talks to the public catalog endpoint over plain HTTP, which is why it is fast and cheap to run.
• Photo URLs are signed by Vinted and expire after some time. Download what you need promptly.

FAQ

Why do I get fewer items than maxItems? Your query genuinely ran out of inventory. The log prints each price slice and its size, so you can see exactly where the catalog ended.

Can I scrape several countries in one run? Not in a single run — pick one marketplace per run. Prices, currencies and inventories differ per country, so mixing them in one dataset would be misleading.

Why are some items repeated in the log counter but not in the dataset? Price slices overlap slightly at their boundaries. Duplicates are removed by listing ID before anything is saved, and the number removed is reported at the end of the run.

Can I get a seller's full inventory? Not through this Actor — it walks the catalog, not member profiles.