Microsoft retired the general Bing Search APIs on August 11, 2025. A “Bing scraper” is therefore not a like-for-like replacement for them. Calling a search API backed by its own index, collecting the results pages a search engine displays, and using Bing’s webmaster tools for your own site are three different things, with different result sets, permitted uses, and reliability. This guide explains what changed, which Bing products still exist, and how to choose between the options as of October 2026.
What changed: the general Bing Search API is retired
Microsoft’s lifecycle notice, published on Microsoft Learn under the title “Bing Search APIs retiring on August 11, 2025” and last updated May 16, 2025, states:
As an Amazon Associate I earn from qualifying purchases.
Bing Search APIs will be retired on August 11, 2025. Any existing instances of Bing Search APIs will be decommissioned completely, and the product will no longer be available to be used or new customer signup.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsSpecial offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.
Code that still calls the retired endpoints should be treated as non-functional, and no new integration should be built on them. For customers who need live web data inside a large language model application, Microsoft points to Grounding with Bing Search in Azure AI Agents. That service is scoped to Azure AI Agents, so confirm its current availability and whether it fits your use before assuming it covers a general results feed.
#1 Best Overall
Bing Webmaster API is a separate product
The retirement does not affect Bing Webmaster API, which continues to serve owners of registered sites. Its documented functions are rank and traffic statistics, links, keywords, and crawl statistics, plus URL and sitemap submission. All of those concern your own registered site, so the tool cannot tell you what Bing shows for an arbitrary query.
One date needs attention. Microsoft’s Bing Webmaster API overview says the legacy SOAP and POX APIs retire on August 31, 2026, and directs users to the REST APIs. That date has now passed. Any tool still calling a SOAP or POX endpoint should be migrated to the REST documentation rather than assumed to work.
Search APIs and SERP scraping are different categories
A search API returns results under the terms of the company that operates it. A SERP collection service or direct scraper gathers what a search engine displays. Those are different categories, and a third-party service’s availability does not establish that the search engine has agreed to it. Compare options on the questions in the table below rather than on the name of the engine.
| Option | What it is | Main question to answer |
|---|---|---|
| Retired Bing Search APIs | Former general-purpose Microsoft search API, retired August 11, 2025 | Not available for new builds; new customer signup has stopped. |
| Bing Webmaster APIs | First-party data and submission tools for sites you have registered | Do you own or have registered the site, and does the REST API cover the metric or submission you need? |
| Grounding with Bing Search | Microsoft’s identified path for giving Azure AI Agents live public-web data | Is it available for your Azure setup, and do its service-specific terms fit your product? |
| Independent-index search API (Brave Search API) | Search API backed by Brave’s own web index; Brave says it does not scrape Google or Bing | Does its index cover your market and language, and do its storage and AI-use restrictions allow your product? |
| SERP collection provider or direct scraping | Collection of results pages from a search engine | Are the provider’s terms and the engine’s terms both compatible with your use, and can you verify the results you receive? |
What the terms restrict
Terms often decide what you can do with results more than the access method does. The points below are service-specific contract language, not a general ruling on scraping.
Microsoft’s Search API terms (retired API)
Microsoft’s terms for the Search API say responses may be used only in internet search experiences, and they prohibit copying or caching data from responses. Review these first if you still hold results captured under that API.
Microsoft’s LLM terms
The LLM terms describe a limited permitted use: grounding a large language model and displaying source attribution. They prohibit using that content to train an LLM. These terms apply to specified tiers, so confirm which tier and service your implementation uses.
Microsoft’s examples of search-like requests
Microsoft’s documentation for the retired API describes acceptable search-like requests as direct user queries or app actions that logically represent a search request. Its examples are:
- a query typed into an app’s search box;
- selected content for which the user asks for more information;
- a user asking a search bot about a specific topic.
Treat these examples as a picture of intended use, not as a route back to the retired API.
Google Programmable Search Engine
Google’s Programmable Search Engine terms prohibit generating queries by automated or invalid means, and they prohibit crawling, spidering, or non-transitory storage of information obtained through the service. Google’s help documentation also describes a transition to more focused solutions beginning January 20, 2026. That date has passed, so confirm what the service currently offers before treating it as an option.
Brave Search API terms
Brave’s API terms, last updated September 1, 2026, prohibit storing or caching results beyond transient operational storage, redistributing results, and using results to create, train, benchmark, or improve AI models or services. The benchmarking and model-improvement language matters if you plan to compare Brave’s output with another provider’s. Brave says its API is based on its own web index rather than on scraping Google or Bing, so its results should not be expected to match a Bing results page.
Why two Bing result pages can differ
Microsoft explains that Bing crawls and indexes pages, then ranks them using factors that include relevance, quality and credibility, engagement, freshness, location, and language. Relative importance can vary, so the same query can return different results depending on when it runs, where it runs, and in what language.
For a scraper, each run is a snapshot of one context. Do not promise customers a stable or universal result set, and when you compare two providers, match the market, language, and time of the query first.
Best Value
Choosing a path
- You need performance data, indexing status, or URL and sitemap submission for a site you own: use Bing Webmaster Tools and its REST API documentation. The retirement of the general Search API does not change this product’s purpose.
- An Azure AI Agent needs live public-web data: evaluate Grounding with Bing Search, confirming current availability and its terms for your exact implementation.
- You need general programmatic web search and can accept a different index: evaluate an independent-index API such as Brave Search API against the restrictions described above.
- You must have exact Bing or Google results: work through the provider checklist below. The sources reviewed do not verify any particular vendor as authorized, reliable, or best value.
Do not present “scrape Bing directly” as an equivalent of the retired API. Microsoft’s retirement is explicit, and a third-party service that remains reachable does not establish that Microsoft or any search engine has granted permission for your use.
Checklist for a SERP data provider
- Source: whether results come from a named search engine, from the provider’s own index, or from a mix, and whether the provider states this in writing.
- Terms: the provider’s terms and the search engine’s terms, read together, with the date of each document noted.
- Permitted use: display, storage, caching, redistribution, and AI use, each checked against what your product actually does.
- Market controls: whether you can set country, language, and location, and whether those controls match your users.
- Freshness and repeatability: how often results refresh, and how much two runs of the same query can differ.
- Reliability: blocking behavior, error handling, and any uptime or accuracy commitment in writing.
- Retention: how long results are kept, and whether you can delete them.
- Cost: the pricing model and what counts as a billable request, taken from the provider’s current pricing page.
- Legal review: the law that applies where you and your users are. Terms are not a legal opinion, and the sources reviewed do not settle how any method is treated in every jurisdiction.
What this guide does not establish
This guide reports documented terms and lifecycle dates. It is not based on hands-on testing of any service. It does not benchmark SERP providers against each other, does not analyze the law of any particular jurisdiction, and does not offer search-volume, accuracy, latency, coverage, or price comparisons, because the sources behind it do not establish them.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Recommended Free Tools
