Skip to content

Run Yandex Reverse Image Search via API

Yandex reverse-image search answers a different question than a keyword search: where else does this picture appear, what does Yandex think it depicts, and does it carry a recognizable brand mark?

OUTPUT FORMATSHTMLJSONMarkdown

Doing that by hand is slow and hard to repeat. An API workflow lets you send an image URL or visual query, let a real browser execute Yandex's reverse-image flow, and receive matching image URLs, source pages, similarity metadata, and classifications back as structured data.

What a reverse-image lookup returns

Each Yandex reverse-image result page can yield matching image URLs, the pages those images appear on, similarity scores, image dimensions, visual classifications, and brand-match data. That combination supports three distinct jobs: tracing where an image is reused, grouping images by visual category, and checking whether a logo appears in unexpected places. The API handles JavaScript rendering, residential proxy rotation, and bot-detection bypass so the request looks like ordinary traffic rather than a bare script.

Three jobs this fits

Source-finding: pass an image URL and collect every page where Yandex reports the image appears. This is useful for rights checks, attribution, and duplicate detection. Category tagging: pull Yandex's visual classifications per match to support trend research or image-grouping pipelines. Brand-logo detection: collect brand-logo matches across the web for brand-protection and counterfeit-detection workflows. Each job uses the same request shape but different extraction fields, so decide which output you need before writing selectors.

Request controls and what they change

The controls below map a request setting to the problem it solves. Use this as a checklist when a lookup returns incomplete or blocked results.

Example: a brand-protection lookup

Suppose a marketplace team wants to know whether a competitor is reusing its product photography with the logo cropped out. The team submits the image URL with render_js=true so Yandex's reverse-image processing completes in a real browser, enables stealth_proxy to reduce the chance of an aggressive block, and supplies ai_extract_rules describing the fields it wants back — matching image URLs, source pages, similarity scores, and brand-match flags. The response arrives as parsed JSON, which the team can compare against its own asset registry. If the target throttles aggressive patterns, the team spaces requests out or uses multiple sessions rather than assuming a per-target rate limit exists.

Telling failures apart

A blocked request, a JavaScript gap, a geo-variance issue, and an extraction gap look similar in a log but need different fixes. If requests are refused or challenged, enable stealth_proxy. If the page returns before results appear, pass render_js=true so the reverse-image flow executes server-side. If results differ by region, set country_code to route through a residential IP in the country you care about. If the page loads but fields are missing, supply extract_rules with CSS selectors mapped to field names, or switch to ai_extract_rules with a natural-language spec. Concurrency limits are per plan, not per target; if Yandex throttles aggressive patterns, space requests out or use multiple sessions.

Output formats beyond JSON

The same request pipeline can return screenshots for visual verification, Markdown for LLM pipelines, or raw HTML when you need to inspect the page yourself. A js_scenario can upload an image, wait for Yandex's reverse-image processing to finish, and scroll the results grid so matches that only appear after the visual search completes are captured. For cross-engine visual discovery, a separate Google Search API can pull Yandex reverse-image references from Google SERPs.

From account to first request

Create an account to receive an API key and free test credits, then install the SDK for your stack (Python, Node.js, Go, PHP, Ruby, or Java) or use raw cURL. Send the Yandex reverse-image URL with your API key as a parameter; the renderer executes the page in a real browser, rotates the IP, and returns a response your scraper can act on. The dashboard request builder lets you paste a URL, set parameters, and copy a generated snippet in any of those languages if you prefer not to write code first.

Request control map
ControlWhat it doesUse it when
render_jsRuns the page in a real headless browser and returns post-render HTMLResults appear only after JavaScript executes
stealth_proxyRoutes the request through stealthier proxiesThe target challenges or blocks ordinary requests
country_codeRoutes through a residential IP in the chosen countryContent varies by geography
extract_rulesCSS selectors mapped to field namesYou know the page structure and want specific fields
ai_extract_rulesNatural-language spec for structured extractionYou want structured JSON without writing selectors
js_scenarioScripted browser actions (upload, wait, scroll)Matches appear only after the visual search completes

Controls reflect the documented request parameters; behavior depends on the target page and plan.

Frequently asked questions

Will I get blocked when scraping Yandex Reverse Image Search?

Requests are routed through rotating residential proxies and a real headless browser so they resemble normal traffic. If a target becomes aggressive, enable stealth_proxy on the request.

Can I scrape Yandex Reverse Image Search without writing code?

Yes. The dashboard request builder lets you paste a URL, set parameters, and copy the generated snippet in Python, Node, Go, PHP, Ruby, or cURL.

How do I handle JavaScript when scraping Yandex Reverse Image Search?

Pass render_js=true. The page runs in a real headless browser, the JavaScript executes server-side, and the post-render HTML is returned.

Can I scrape Yandex Reverse Image Search from a specific country?

Yes. Pass country_code to route through a residential IP in that country. This matters when content varies by geography.

What's the rate limit for scraping Yandex Reverse Image Search?

Concurrency limits apply per plan, not per target. If the target throttles aggressive patterns, space requests out or use multiple sessions.

How do I get structured JSON from Yandex Reverse Image Search?

You have two routes to parsed JSON. The first is extract_rules: you map CSS selectors to the field names you want, and the response is returned as JSON keyed by those names. The second is ai_extract_rules: instead of selectors, you describe the fields in plain language, and the same parsed-JSON response comes back. Pick selectors when the page structure is stable and you want tight control over each field; pick the natural-language spec when the layout shifts or you are still deciding which fields matter.