Yandex Search API

GET | POST /v1/yandex/search

Send parameters as a query string with GET, or as a JSON body with POST. Both are equivalent.

Request

curl "https://gateway.apinoa.com/v1/yandex/search?q=moscow+kremlin&type=images" \
  -H "x-api-key: $APINOA_API_KEY"

Parameters

Parameter	Type	Notes
`country`	enum	The market, as an ISO 3166 country code. `US` `RU` `TR` `KZ` `BY` `UA` `UZ` `DE` `GB` `FR`
`cursor`	string	The previous page's `cursor`. Pass it alone for the next page.
`device`	enum	Which page to read: the desktop (default) or the mobile one. `desktop` `mobile`
`image`	string	An image URL: a reverse-image search instead of `q` (`type=image`).
`lang`	enum	Interface language: en (yandex.com, the default) or ru (yandex.ru). `en` `ru`
`limit`	integer 1–100	Results wanted (default 20, at most 100). More than one page is charged per page.
`q`	string	The query, search operators included.
`timeRange`	enum	Only results from this past period. `hour` `day` `week` `month` `year`
`type`	enum	What to search: news, images, videos, or `image` for a reverse-image search. Yandex's web results are not offered. `news` `images` `videos` `image`

The same parameters, and every response type, are in the OpenAPI document.

Response

Returns ResultsPage in data. A real response, shortened: arrays keep their first entries and long text is cut.

{
  "success": true,
  "data": {
    "engine": "yandex",
    "query": "moscow kremlin",
    "type": "images",
    "outcome": "results",
    "country": "US",
    "lang": "en",
    "device": "desktop",
    "page": 1,
    "items": [
      {
        "type": "image",
        "imageUrl": "https://upload.wikimedia.org/wikipedia/commons/thumb/c/c1/DSC07437-%D0%9C%D0%BE%D1%81%D0%BA%D0%BE%D0%B2%D1%81%D0%BA%D0%B8%D0%B9_%D0%9A%D1%80%D0%B5%D0%BC%D0%BB%D…",
        "thumbnail": "https://avatars.mds.yandex.net/i?id=0478d2b2c1e3bc635da290a7ec93ed7a8389541d-4568991-images-thumbs&n=13",
        "width": 1280,
        "height": 854,
        "title": "Кремль - Википедия",
        "url": "https://ru.wikipedia.org/wiki/%D0%9A%D1%80%D0%B5%D0%BC%D0%BD%D0%B8%D0%BA",
        "source": "ru.wikipedia.org",
        "ext": {
          "id": "c13a461cad3531a9d14e15d194b1f453",
          "snippet": "Кремль — Википедия",
          "copies": [
            {
              "url": "https://day.ru/public/styles/1920x1080_16_9_watermark/public/image/2026-07/screenshot_8.jpg.webp?itok=tgB7-t6o",
              "width": 1920,
              "height": 1080,
              "bytes": 239798
            },
            {
              "url": "https://apostrophe.ua/uploads/image/a2d88e4418642f7271777de4cf54216f.jpg",
              "width": 1500,
              "height": 1001,
              "bytes": 1802215
            }
          ]
        },
        "rank": 1
      },
      {
        "type": "image",
        "imageUrl": "https://cdn.culture.ru/c/367836.jpg",
        "thumbnail": "https://avatars.mds.yandex.net/i?id=8b172529aea8e52da70a8539d9ee647e8646aea7-5338433-images-thumbs&n=13",
        "width": 1350,
        "height": 900,
        "title": "Музеи Московского Кремля: история основания, экспозиции, соборы на территории Кр",
        "url": "https://www.culture.ru/materials/94949/muzei-moskovskogo-kremlya-chto-i-gde?7",
        "source": "culture.ru",
        "ext": {
          "id": "e07ff0761619f4d6f544159823e1163d",
          "snippet": "Telegram.",
          "copies": [
            {"url": "https://cdn.culture.ru/c/367836.jpg", "width": 1350, "height": 900, "bytes": 152243},
            {
              "url": "https://moscowrussian.com/images/kremlin-church.jpg",
              "width": 1200,
              "height": 800,
              "bytes": 263551
            }
          ]
        },
        "rank": 2
      },
      {
        "type": "image",
        "imageUrl": "https://news.store.rambler.ru/img/b4d984f9a02f4f2f79bdc5a25fec1378?img-format=avif&img-1-resize=width:1280",
        "thumbnail": "https://avatars.mds.yandex.net/i?id=2a6a772473ef11a574759d3c2ef99700d60d3073-12935885-images-thumbs&n=13",
        "width": 1280,
        "height": 720,
        "title": "Что посмотреть в Московском Кремле: краткая история, достопримечательности и цен",
        "url": "https://weekend.rambler.ru/tourism/56962355-chto-posmotret-v-moskovskom-kremle-kratkaya-istoriya-dostoprimechatelnosti-i-tseny-na-ekskursii/",
        "source": "weekend.rambler.ru",
        "ext": {
          "id": "19300ddc2386aec2e7a879ec0a199c96",
          "snippet": "© yulenochekk/iStock.com",
          "copies": [
            {
              "url": "https://www.his-j.com/oe/up_img/LED0097/Tr00100420201016215043586.jpg",
              "width": 4774,
              "height": 3106,
              "bytes": 5002584
            },
            {
              "url": "https://cdn1.img.spnfa.ir/img/03/76/037659_0:139:3147:1909_1920x0_80_0_0_8aabe5681b4393b85463dc85a65e816a.jpg",
              "width": 1920,
              "height": 1080,
              "bytes": 373184
            }
          ]
        },
        "rank": 3
      }
    ],
    "fetchedAt": "2026-10-30T01:00:16.470Z"
  }
}

Errors

Status	Code	Meaning
400	`INVALID_CURSOR`	The `cursor` is not one this operation returned. Start again without it.
400	`INVALID_PARAMS`	A parameter is missing, unknown or out of range, or the engine does not take it. The message names it.
401	`UNAUTHORIZED`	The `x-api-key` header is missing, or the key is revoked or unknown.
402	`INSUFFICIENT_BALANCE`	The balance does not cover this call. Top up in the dashboard.
429	`MARKETPLACE_SATURATED`	Too many concurrent calls. Retry after the `Retry-After` seconds.
502	`UPSTREAM_ERROR`	The engine answered with something unusable. Retry with backoff.
503	`SERVICE_UNAVAILABLE`	A temporary failure. Retry with backoff.

Every code is listed in Errors. Failed calls are not billed.

Pricing

Price $0.0012 per call, plus $0.0012 for each extra call it makes — at most 4 more, so $0.006 in total. List price; the rate steps down above 1,000 calls a day.

Notes

Pass type: Yandex's web results are not offered. News comes from Yandex's Russian news edition and takes no country or lang.
type=image with an image URL instead of q runs a reverse-image search.
limit above one page reads further pages, one unit each.

Search