Avito Scraper

Avito Scraper (Apify Actor)

Scrapes search-result listings from Avito (avito.ru). You provide a keyword, optional category/region and a price range; the actor builds the search URL, paginates through results and stores each listing card in the dataset.

Scope: search-result cards only. It does not open individual ad pages and does not collect phone numbers.

⚠️ Proxies required. Avito is geofenced and Cloudflare-protected — you need a large Russian residential (or mobile) IP pool (prefilled to Apify Residential, country RU). Without one the run collects nothing. See Proxies.

Features

• Structured search — by keyword, region, category and price range, with sorting.
• Or paste URLs — feed ready-made Avito search/category URLs via startUrls.
• HTTP-only engine — Chrome-TLS-fingerprinted HTTP (impit) does everything: solves Avito's firewall proof-of-work challenge in-process, follows the keyword→category redirect and paginates. A headless browser is kept only as a last-resort fallback.
• Rich data from Avito's own JSON — exact price, posted date, GPS coordinates, category, photo count and seller rating; robust to HTML/markup changes.
• Public data only — listing cards, no phone numbers and no personal data.

How it works

Avito guards its pages with two layers, and the actor handles both over plain HTTP:

1. TLS fingerprinting — plain Node HTTP clients are rejected. The actor uses impit (real Chrome TLS + HTTP/2 fingerprint, no browser) through CheerioCrawler.
2. Firewall proof-of-work challenge (HTTP 439, «Нужна небольшая проверка браузера») — a small SHA-256 puzzle normally solved by the page's JavaScript. The actor solves it in-process on the same session/IP (a few milliseconds), re-fetches, and the session stays unblocked for ~10 minutes.

A plain keyword search (/{region}?q=...) returns a shell whose embedded state names the canonical category URL Avito would redirect to (/all/{category}/…?context=…). The actor follows that URL over HTTP; it is server-side rendered with the cards and Avito's total count, so the remaining pages are fanned out in parallel and fetched over HTTP too. Ready-made URLs from startUrls go through the same flow.

Listings are parsed from the page's embedded catalog JSON (Avito's server-rendered state), which is more robust than HTML scraping and carries richer fields (exact price, posted timestamp, coordinates, category, photo count, seller rating). DOM selectors remain as a fallback.

The crawler rotates identities via Crawlee's SessionPool, retires blocked sessions (429 / hCaptcha pages, dead proxy IPs) and retries generously — a failed attempt costs ~1–2 s, so a bad IP is just skipped. Only if a start URL cannot be resolved over HTTP at all does a headless Playwright browser step in as the last resort; the resolved URL it finds is handed back to the HTTP pass.

Input

Provide a keyword or startUrls. A bare run with no input still works — it defaults to keyword: "iphone 15" and Apify Residential (country RU) proxy.

Field	Type	Required	Description
keyword	string	no	Search query (the q parameter). Defaults to iphone 15 if neither keyword nor startUrls is given.
startUrls	array	no	Ready-made Avito search URLs (copy from your browser). Take precedence over keyword/region/category and bypass Avito's category redirect.
region	string	no	Region slug from the Avito URL. Default rossiya (all Russia).
category	string	no	Category slug. Empty = all categories.
priceMin	integer	no	Lower bound of the price range (→ pmin).
priceMax	integer	no	Upper bound of the price range (→ pmax).
maxItems	integer	no	Max listings to collect. Default 100.
sort	string	no	Result order: default, date (newest), priceAsc (cheaper first), priceDesc (expensive first). Maps to Avito's s param.
proxyConfiguration	object	no*	Proxy. Defaults to Apify Residential, country RU.

* A large Russian residential (or mobile) proxy pool is effectively required — without one the run collects nothing. See Proxies.

Region / category slugs

Use the slugs exactly as they appear in an Avito URL, e.g.:

• Regions: rossiya, moskva, sankt-peterburg, novosibirsk, ekaterinburg
• Categories: avtomobili, kvartiry, telefony, noutbuki, tovary_dlya_kompyutera

Tip: open Avito in a browser, set your filters, and copy the path segments from the resulting URL.

Example input

{
  "keyword": "iphone 15",
  "region": "moskva",
  "category": "telefony",
  "priceMin": 30000,
  "priceMax": 70000,
  "maxItems": 100,
  "proxyConfiguration": {
    "useApifyProxy": true,
    "apifyProxyGroups": ["RESIDENTIAL"],
    "apifyProxyCountry": "RU"
  }
}

Output

Each dataset item (fields are read from Avito's embedded catalog JSON):

{
  "id": "7685094754",
  "title": "iPhone 17, 256 ГБ",
  "price": 53490,
  "priceString": "53 490 ₽",
  "currency": "RUB",
  "url": "https://www.avito.ru/ekaterinburg/telefony/iphone_17_256_gb_7685094754",
  "location": "Свердловская обл., Екатеринбург",
  "locationId": 654070,
  "lat": 56.830664,
  "lng": 60.589794,
  "categoryId": 84,
  "categorySlug": "telefony",
  "categoryName": "Телефоны",
  "postedDate": "2026-05-31T20:05:23.000Z",
  "postedTimestamp": 1780257923000,
  "imageUrl": "https://...",
  "photoCount": 7,
  "description": "iPhone 17, 256 ГБ. Оригинальный, запечатанный…",
  "sellerRating": 4.9,
  "sellerReviews": "3 252 отзыва",
  "isVerified": false,
  "isNew": false,
  "isPromoted": false,
  "category": "telefony",
  "region": "rossiya",
  "keyword": "iphone",
  "scrapedAt": "2026-05-31T12:00:00.000Z"
}

category/region/keyword echo your search input; categorySlug/locationId/etc. describe the listing itself. Some fields can be empty for certain listings (e.g. location/lat/lng when Avito hides the exact address — locationId is always present). If the embedded JSON is ever unavailable, the DOM fallback fills the core fields (id, title, price, url, location, imageUrl).

Proxies — important

Avito sits behind Cloudflare and rate-limits per IP. The TLS impersonation above defeats fingerprint-based detection, but IP reputation and per-IP rate limits still matter — so the proxy pool is decisive.

• Apify Residential, country RU — the prefilled default. Apify Proxy has no selectable "mobile" group, so residential is the strongest Apify-native option for Avito (its pool does include some mobile IPs, but they can't be isolated).
• The pool must be large. A handful of IPs gets detected and banned quickly; Avito's limits are aggressive. Prefer a big rotating residential pool.
• Custom proxies for true mobile IPs. Carrier (mobile) IPs are rate-limited by Avito far less aggressively, but Apify doesn't offer them as a group. To use them, pick "Custom proxies" in the proxy editor and paste URLs from an external mobile provider (e.g. BrightData mobile). Datacenter IPs can work only with a very large rotating pool.

Without a decent Russian IP pool the actor will mostly hit "Доступ ограничен" block pages regardless of the engine.

Running locally

npm install
npx playwright install chromium   # only needed for the last-resort browser fallback
apify login                       # required to use Apify residential proxies locally
apify run --purge                 # reads input from storage/key_value_stores/default/INPUT.json

Results land in storage/datasets/default/.

Deploy

$apify push

FAQ

Why are Russian proxies required? Avito is geofenced and Cloudflare-protected; without a Russian IP pool you mostly get "Доступ ограничен" block pages. See Proxies above.

Does it launch a browser? Normally no — the firewall challenge, the category redirect and pagination are all handled over HTTP. A headless browser is started only as a last resort when a start URL could not be resolved over HTTP at all.

Does it collect personal data? No. Only public listing fields (title, price, location, photos, rating, etc.). No phone numbers and no seller contact details.

Which regions and categories are supported? Any — use the slugs from an Avito URL (see above). Default region is rossiya (all Russia), all categories.

Why are some location fields empty? Avito hides the exact address for certain listings; locationId is always present.

Legal note

Avito's Terms of Service restrict automated scraping. This actor collects only public listing fields and no personal data (no phone numbers, no seller contact details). You are responsible for ensuring your use complies with Avito's terms and applicable law; keep request volumes modest.