What's New

New updates and improvements to SearchApi

← Back

July 2026: Update

New
Today, we're thrilled to update you on some of the newest enhancements and give you a sneak peek into what’s coming next.

GDPR Compliant



SearchApi is GDPR compliant. Our data protection program covers how we collect, process, store, and delete personal data, along with subprocessor management, data subject request handling, retention limits, and breach notification.

If you're evaluating SearchApi or running a vendor security review, our GDPR documentation sits alongside our ISO/IEC 27001:2022 certificate in our ​Trust Center​.

​ChatGPT API​


The ​ChatGPT API​ returns ChatGPT’s answer for any query in formatted JSON. It includes:

  • Text Blocks - The text_blocks array breaks the answer into typed blocks: paragraph, header, ordered_list, unordered_list, table, and code, each able to carry reference_indexes pointing at its sources.
  • Markdown - A markdown field renders the full answer as formatted text.
  • Reference Links - The reference_links array lists each cited source with index, title, link, snippet, source, date, and ref_id.
  • Web Results - With web_search enabled, a web_results array returns the pages ChatGPT browsed, alongside the search_queries it issued.
  • Entity Cards - The expand_entities parameter resolves the cards array into full entity detail (category, disambiguation, sections, related_entities, inline_images).

We kept keys the same as in the Google AI Overview, Google AI Mode, and Bing Copilot APIs for easier migration.

​Gemini API​


The ​Gemini API​ returns Google Gemini’s answer for a query. It includes:

  • Text Blocks - The same typed text_blocks structure, with reference_indexes on each block, plus a full markdown rendering of the answer.
  • Grounding Sources - The reference_links array lists each grounded source with index, title, link, snippet, source, favicon, and rank.
  • Inline Images - An inline_images array recovers the images Gemini embeds in its answer, with title, link, source, original, and thumbnail.
  • Structured Cards - A cards array returns typed answer modules for weather and finance.
  • Response Metadata - A response_metadata object exposes the answering model, language, and a conversation_token, plus a related_questions array for follow-ups.

We kept keys the same as in our other AI-answer engines for easier migration.


​Perplexity API​


The ​Perplexity API​ returns Perplexity’s cited answer for a query. It includes:

  • Text Blocks - The same typed text_blocks structure, with reference_indexes on each block, plus a full markdown rendering of the answer.
  • Reference Links - Each source in reference_links carries index, title, link, snippet, source, date, thumbnail, images, and a trust signal.
  • Source Control - The sources parameter selects which corpora Perplexity searches.
  • Structured Cards - A cards array returns typed answer modules, including weather (location, current, forecast) and finance (symbol, price, change_percent, market_cap, pe_ratio, eps, dividend_yield).
  • Related Questions - A related_questions array for follow-up queries.

We kept keys the same as in our other AI-answer engines for easier migration.

​TikTok Profile Videos API​


The ​TikTok Profile Videos API​ returns a public TikTok account’s video feed by username. It includes:

  • Profile - A profile object with id, link, bio, avatar, is_verified, is_private, followers, following, hearts, and posts.
  • Videos - The videos array provides caption, link, iso_date, duration, mentions, is_pinned, is_ad, and the full engagement set (views, likes, comments, shares, saves).
  • Media Detail - Each video carries video_meta (definition, ratio, size, bitrate, fps), music (title, author, cover, duration), and subtitles per language.
  • Ordering & Pagination - sort_by and next_page_token walk the full feed.

​Apple App Store Top Grossing API​



The ​Apple App Store Top Grossing API​ returns Apple’s top-grossing app rankings for a country. It includes:

  • Top Grossing Apps - The top_grossing_apps array provides position, id, bundle_id, title, link, thumbnail, logos, developer, category, and release_date.
  • Pricing - Each app carries a price object with extracted_price, currency, currency_symbol, and is_free.
  • Targeting - category, store_device, and num scope the chart returned.

We kept the structure consistent with the Apple App Store Top Charts API for easier migration.

​MCP: OAuth 2.1 and the Web Search Bundle​



Our MCP server now supports OAuth 2.1 authentication so that agent clients can connect without a hardcoded API key. We also shipped a Web Search tool bundle grouping our core search engines into a single toolset, and added tools for the new AI-answer and TikTok engines.


⚙️ Engine Improvements:


Google

  • Added link and product_token to inline_shopping
  • Added new ad parsers, ad badges, and sitelink tracking links
  • Fixed ads sitelinks on desktop and mobile
  • Fixed local website links
  • Fixed top_stories in the new news item layout
  • Added experiences to the knowledge graph
  • Empty pages now return a no-results error alongside dmca_messages instead of the messages alone

Google AI Mode

  • Added gl and hl parameters
  • Fixed parsing for the layout changes

Google Finance

  • Fixed the markets section
  • Fixed empty results on Google’s updated page layout

Google Flights

  • Added the expanded_search parameter. Combined with show_hidden_flights or show_cheapest_flights, it might return the complete flight list instead of the first-page subset at the cost of speed (some searches go from 104 to 150+ flights)

Google Lens

  • Fixed blurred exact_matches with safe_search=off

Google Maps Directions

  • Transit buy_ticket options now include their link on link-only layouts
  • Fixed booking links

Google News

  • Fixed organic link extraction for the new layout variant

Google News Portal

  • Fixed article links
  • Fixed sub_menu_sections dropping out on topic pages

Google Play Product

  • Fixed reviews, TV, and audiobook parsers
  • Added review filter parameters

Google Product

  • Added new key related_products: each item’s product_id returns the catalog id accepted by the Google Product Page API, alongside new product_token, position, and currency
  • Fixed an issue with an empty Google Product response

Google Product Offers

  • Fixed an issue with an empty Google Product Offers response

Google Rank Tracking

  • Fixed an issue when not all results were returned

Google Scholar

  • Fixed total_results

Google Travel Explore

  • Fixed bounding-box corners
  • Fixed swapped bounding-box coordinates that returned zero destinations

Airbnb Property

  • Added taxes and extracted_taxes

Airbnb Search

  • Added total_with_taxes and extracted_total_with_taxes for tax-inclusive pricing
  • Fixed the price breakdown to carry every priced line with signed amounts, and discount on regulated-total cards
  • Added has_free_cancellation, and folded bathrooms into accommodations

Amazon Product

  • Fixed missing images for products with apostrophes in their titles

Bing

  • Added time_period, time_period_min, and time_period_max date filters
  • Fixed the organic result domain
  • Fixed ads extraction in the new layout

Bing Copilot

  • Fixed missing reference_links

eBay Search

  • Deprecated the sold_listings and completed_listings filters
  • Fixed search organic results for the new item card layout

Facebook Business Page Reviews

  • Added page, rating, and reviews_count

LinkedIn Ad Library

  • Fixed next_page_token pagination
  • Fixed empty first pages
  • Added parsing for employer-brand ads

Meta Ad Library

  • Added the search_type parameter (keyword_unordered, keyword_exact_phrase)
  • Fixed exact-phrase search with multiple quoted phrases

Walmart Product

  • Fixed live products intermittently returning empty results, with a specific “Walmart couldn’t find this product.” error for products that truly don’t exist
  • Unresolvable product IDs now return product-not-found instead of a generic empty result

Walmart Search

  • Closed store_id values now return a clear error right away instead of timing out

Yandex

  • Pagination past page 4 is supported. The stale limit has been removed from the docs

Yandex Reverse Image

  • Fixed handling of image URLs with query parameters

Zillow Property

  • Fixed special_offers
  • Fixed “no property data” errors

Zillow Search

  • Added amenity with rarity and extracted_rarity

⏭️ What's coming next?


  • GitHub APIs
  • Google Flight Deals API
  • Booking API
  • Microsoft Copilot API
  • TikTok Search API
  • TikTok Video API
  • TikTok Transcripts API

Stay tuned for more updates! As always, we're committed to providing you with the best experience.