What Apple Photos already does well
Apple Photos is on the phone already, costs nothing, and is where Apple’s own indexing work surfaces first: the People & Pets collection it builds from your library, search by who added a picture to an iCloud Shared Photo Library, and the text Apple has recognised inside your images. For a large share of everyday searches it is simply the right tool, and nothing below is meant to talk you out of it.
Apple’s own iPhone User Guide lists what the Photos search field accepts, and the list is longer than most people realise.
- Date (month or year), place (city or state), business names such as museums, categories such as “beach or sunset”, and events such as sports games or concerts.
- People or pets you have named — Photos “sorts photos and videos of people, dogs, and cats in the People & Pets collection”, and Apple states that this face analysis happens on the device.
- Text inside a photo, “an email address or phone number, for example”, plus any caption you have written.
- With Apple Intelligence, a plain-language description: Apple’s own example query is “Maya skateboarding in a tie-dye shirt.” Apple says you can also “search video clips for key moments.”
- Duplicates, under Collections › Utilities › Duplicates, where a set is combined with Merge.
- One library across devices: iCloud Photos keeps the pictures and the People & Pets names in sync, so the work of naming someone is done once. Queryable has no equivalent — its index is local to one device.
Compare
Every Apple cell below is a fact Apple publishes, with the source numbered at the end of the cell. Where Apple publishes nothing, the cell says so rather than guessing. Every Queryable cell describes the shipping app.
| Dimension | Apple Photos (iOS 27) | Queryable 2.1.6 |
|---|---|---|
| Price | Free. Photos is part of iOS; there is nothing to buy and no subscription. | $4.99 in the US App Store, paid once. No subscription, no in-app purchases, no free tier and no trial. In mainland China, Hong Kong, Macao and Taiwan the app is sold separately as 寻隐 / 尋隱, at ¥1 on the mainland China storefront. [7] |
| Devices and Apple Intelligence | Photos search itself is part of iOS. Description search is an Apple Intelligence feature, and Apple lists Apple Intelligence on iOS 27 as available on “iPhone 16 models or later, iPhone 15 Pro, iPhone 15 Pro Max, iPhone Air, iPad mini (A17 Pro), … iPad models with M1 or later”. Apple also lists “up to 8 GB of storage” on those models — up to 14 GB on the newest ones — and requires the device language and Siri language to be set to the same supported language. [1][2] | Any iPhone or iPad running iOS 18.0 or later. No Apple Intelligence, no account, no sign-in, and no feature gated behind a newer chip. [7][8] |
| Where the processing happens | Apple says Photos “uses on-device machine learning” and that its on-device analysis “includes scene classification, people and pets identification, photo quality analysis, and audio classification”. Enhanced Visual Search is the documented exception: “Your device privately matches places in your photos to a global index Apple maintains on our servers,” protected by homomorphic encryption, differential privacy and an OHTTP relay, and switchable off in Settings. Apple does not publish where the Apple Intelligence description search itself runs. [3][1] | On the device. Indexing and search run through Core ML and Metal on the phone, the app makes no network request of its own, and it ships no analytics or crash-reporting SDK. [8][9] |
| Describe a scene in words | Yes, with Apple Intelligence. Apple: “With Apple Intelligence, you can search for specific images by describing what you want to find. You can also search video clips for key moments.” The guide keeps it conditional throughout — “If you’re using Apple Intelligence, enter a description using natural language” — and lists the ordinary search terms separately. [1] | Yes, on every supported device. A sentence is encoded on the phone by a MobileCLIP-S2 text model into the same 512-dimension space as the photos, and the library is ranked by similarity. Searching video by its moments is not a Queryable feature. [8] |
| Search with an example photo | Apple does not publish this. The search behaviour Apple documents is typed text and tapped suggestions; the guide describes no way to search the library by supplying a reference image. [1] | Yes. A camera button in the search bar either takes a photo or picks one from the library and uses that image as the query, and any photo’s detail view has a “Similar Photos” panel built from the embedding already stored for it. [8] |
| Text inside photos (OCR) | Yes. Apple lists “Text (an email address or phone number, for example)” among the things the Photos search field accepts, and Live Text separately reads text in the camera frame — though Apple notes Live Text “isn’t available in all regions or languages”. [1][6] | No. Queryable performs no text recognition of any kind. A screenshot is found by what the image looks like, never by the words printed in it. [8] |
| People and pets | Yes. Photos “sorts photos and videos of people, dogs, and cats in the People & Pets collection”; you name a person, dog or cat once and then search by that name. Apple states the recognition is on-device, and names stay current across iCloud-enabled devices. [4][3] | No face, people or pet recognition at all — and no albums, no location and no map. Naming someone is not possible. [8] |
| Time filtering and sorting | Yes. A date — month or year — is one of the documented search terms, and search results can be sorted by “Added” or “Captured” and filtered by options such as Favorites or Edited. [1] | Time filters only: Last Week, Last Month, 3 Months, 6 Months, Last Year, or a specific year taken from the range of photos actually indexed. Results are ranked by similarity and cannot be sorted by date. [8] |
| Duplicate and similar-photo cleanup | Duplicates only. “The Photos app identifies duplicate photos and videos in your photo library in the Duplicates collection,” under Collections › Utilities, where a set is combined with Merge. If the library has none, the collection does not appear. Apple does not publish a near-duplicate or similar-scene grouping. [5] | Three tiers from a single scan — Exact Duplicates, Nearly Identical and Similar Scenes — with an aesthetic score pre-selecting the photo to keep, and one system confirmation dialog for the whole batch. Stills only: videos are never grouped and never deleted. [8] |
| Query languages and regions | Apple Intelligence “is available with Siri and device language set to Chinese (Simplified), Chinese (Traditional), Danish, Dutch, English, French, German, Italian, Japanese, Korean, Norwegian, Portuguese, Spanish, Swedish, Turkish, or Vietnamese,” and the device and Siri language must match. Apple also states that Apple Intelligence features “will not currently work for supported devices purchased in China mainland”. [1][2] | Type in your own language. The English edition translates a non-English query on the device with Apple’s Translation framework and searches with the original and the translation together, keeping whichever matches a photo better; that translation needs a one-time free download of Apple’s language pack. The interface ships in 18 languages. In mainland China, Hong Kong, Macao and Taiwan the app is sold as 寻隐 / 尋隱, which has its own natively trained Chinese text encoder and no translation step. [8] |
| Offline behaviour | Apple does not publish this. The Photos search documentation does not state whether search works with no network connection. Apple does state that Enhanced Visual Search matches places in your photos against an index held on Apple’s servers. [1][3] | Once a photo is indexed, searching for it works with networking switched off. Two boundaries: the one-time language-pack download above, and iCloud-only originals — indexing runs with network access disabled, so a photo with no usable copy on the device is stored as a placeholder and is not retried. Make those photos available locally in Apple Photos before building the index. [8][10] |
| Open source | No. Apple does not publish the source code of Photos. | Partly, and the caveat matters. The MIT-licensed public repository mazzzystar/Queryable holds the CLIP indexing and search core as it stood on 29 March 2026. Everything added in 2.1.2 to 2.1.6 — translation-based search, search by photo, time filters, the three-tier cleanup — is developed in a private repository and is not in the public one. [9] |
| Result quality, accuracy and speed | Not tested. | Not tested. |
Result quality, accuracy and speed: No ranking quality, recall, precision, indexing speed or search latency was measured for this page. A claim of that kind requires the controlled hands-on test described in the project’s comparison research protocol, and this page is deliberately published without one.
When Apple Photos is enough
If all of the following are true, Apple Photos will do what you need and a second app is an unnecessary purchase.
- Your iPhone is on Apple’s list of Apple Intelligence-enabled models, your device and Siri language is one of the sixteen Apple supports, and your device was not purchased in mainland China — Apple states Apple Intelligence features “will not currently work for supported devices purchased in China mainland”.
- You mostly search for a named person or pet, a place, a date, or the text printed in a screenshot or receipt — all of which Apple Photos does and Queryable does not.
- You want the same search on iPhone, iPad and Mac without indexing each device separately.
- You would rather not pay for anything.
When Queryable adds something
Queryable is worth its one-time price in a narrower set of situations. They are mostly situations where Apple’s description search is unavailable to you, or where the thing you want to do is not searching by words at all.
- Your iPhone or iPad is not on Apple’s list — which leaves out every iPhone before the 15 Pro, and the standard iPhone 15 and 15 Plus — but it runs iOS 18.0 or later. Description search then behaves the same as it does on the newest phone.
- Your device language is not one of Apple’s sixteen, or your device was purchased in mainland China, where Apple states Apple Intelligence features do not currently work. The Greater China edition, 寻隐, searches Chinese with a natively trained Chinese encoder rather than a translation step.
- You want search that makes no network request at all: the app issues none of its own, contains no analytics SDK, and keeps working in airplane mode.
- You want to search with a picture instead of with words, or pull up “similar photos” from any result you are already looking at.
- You want finer cleanup than “duplicates” — near-identical frames and whole similar scenes, with the keeper chosen for you and videos left untouched.
- You want to read the indexing and search code.
What this page does not claim
This is a facts-only comparison. Nothing here was measured, and the honest list of what Queryable cannot do is longer than the list of what Apple Photos cannot do.
- Neither app’s result quality was tested. No sentence on this page says one app finds photos better, faster or more accurately than the other.
- Apple does not publish whether its description search runs entirely on the device, or whether Photos search works with no network connection. Those rows say so rather than guessing.
- Queryable does not read text in photos, recognise faces, sort results by date, open a result in Apple Photos, or appear in Spotlight, Shortcuts or widgets.
- Queryable indexes only while its indexing screen is open. There is no background or automatic indexing; new photos are added when you tap to add them.
- Queryable reads the system photo library. Photos synced onto the device from a Mac into custom albums are outside what it can see.
- The index lives on one device and is not synced or backed up. Two devices means indexing twice.
- The public repository lags the shipping app by several releases.
- Queryable publishes this page. Read it as our argument, checked against Apple’s documents, rather than as a neutral review.
How this page was compiled
Every Apple figure on this page comes from Apple’s own pages — the iPhone User Guide for iOS 27, Apple’s Apple Intelligence availability article, and the Photos & Privacy page on Apple Legal — read on 20 September 2026 and quoted rather than paraphrased wherever the wording matters. Every Queryable figure comes from the shipping 2.1.6 build and the live App Store listing, checked on 19 September 2026. Prices, device lists, language lists and regional availability change without notice: where this page and Apple disagree, believe Apple.
What was tested: nothing. This is a facts-only comparison assembled from published documentation, and it is published under the Tier 1 rules of the project’s comparison research protocol, which allow a comparison page without a hands-on test provided every cell is a cited vendor fact, a verified Queryable fact, or the words “not tested”. No row states a ranking, accuracy, recall or speed figure, and the comparison table carries an explicit “not tested” row for result quality.
Storefront and versions: Apple Photos as documented for iOS 27 in the United States English iPhone User Guide; Queryable 2.1.6, released 11 August 2026, US App Store, minimum iOS 18.0. The Greater China edition 寻隐 / 尋隱 shares the same version number and the same release track. Corrections are welcome through the project’s GitHub repository.
Questions about Apple Photos and Queryable
Do I still need Queryable if my iPhone has Apple Intelligence?
Often not. If your iPhone runs Apple Intelligence in a supported language, Apple Photos already searches from a description, and for many people that is the whole reason they would have bought a second app. Queryable adds three things Apple does not document: searching with an example photo, a three-tier similar-photo cleanup, and a search path that makes no network request of its own. It also runs on every iPhone and iPad with iOS 18.0 or later, including the models Apple Intelligence does not support.
Does Apple Photos search work without Apple Intelligence?
Yes. Apple’s iPhone User Guide says the Photos search field accepts a date, a place, a business name, a category such as “beach or sunset”, an event, people or pets you have named, text such as an email address or phone number, and captions you have written. Apple presents searching by a free-form description separately, as an Apple Intelligence feature.
Can Queryable find the text inside a screenshot?
No. Queryable performs no text recognition, so it can only match a screenshot by what the image looks like, not by the words printed in it. Apple Photos does search text inside photos — Apple lists “Text (an email address or phone number, for example)” among its search terms — so for receipts, signs, screenshots and documents, Apple Photos is the right tool.
Which one works with no internet connection?
Queryable, with two caveats. Once a photo has been indexed, searching for it runs entirely on the device and keeps working in airplane mode; but the English edition’s translation of non-English queries needs a one-time free download of Apple’s language pack, and photos stored only in iCloud must be made available on the device in Apple Photos before Queryable can index them. Apple does not publish whether Photos search works with no network connection; Apple does state that Enhanced Visual Search matches places in your photos against a global index held on Apple’s servers, and that it can be turned off in Settings.
Which one finds photos of a specific person?
Apple Photos. It groups faces into a People & Pets collection, lets you name a person, dog or cat, and then finds them by that name, and Apple states the recognition happens on the device. Queryable has no face, people or pet recognition at all, so a person’s name is not something it can search for.
Can I use both?
Yes, and that is the normal arrangement. Both read the same photo library through Apple’s Photos framework and neither replaces the other: Apple Photos for people, places, dates and the text inside images; Queryable for describing a scene in a sentence, searching with an example photo, and clearing out near-duplicates.
Sources
Every Apple statement above was read on this site’s behalf from the pages below on 20 September 2026. Apple’s documentation changes without notice; if a page now says something different, Apple is right and this one is stale.
- Search for photos and videos on iPhone — iPhone User Guide (iOS 27) Source for the list of accepted search terms, the Apple Intelligence description-search sentence and its example query, the result sort and filter options, and the Apple Intelligence device and language footnote.
- How to get the next generation of Apple Intelligence Source for the supported device list, the storage and matched-language requirements, the sixteen supported languages, and the mainland China statement.
- Photos & Privacy (Apple Legal, page dated 12 December 2025) Source for on-device analysis in Photos, on-device face recognition, and the Enhanced Visual Search server index with homomorphic encryption and an OHTTP relay.
- Find and name people and pets in Photos on iPhone — iPhone User Guide Source for the People & Pets collection, naming a person, dog or cat, and iCloud sync of those names.
- Merge duplicate photos and videos on iPhone — iPhone User Guide Source for the Duplicates collection under Collections › Utilities and the Merge action.
- Use Live Text with your iPhone camera — iPhone User Guide Source for Live Text reading text in the camera frame, and for its regional and language limits.
- Queryable on the App Store Source for the one-time price, the absence of in-app purchases, the minimum iOS version and the supported device families.
- How Queryable works The on-device model, the local index, the search path and the app’s documented limits, taken from the shipping 2.1.6 build.
- mazzzystar/Queryable — MIT licence The public indexing and search core. Last pushed 29 March 2026, so it predates the 2.1.2–2.1.6 features listed above.
- Offline AI photo search The iCloud-only boundary and what “offline” does and does not cover during indexing.