Where Ente Photos wins
Ente Photos is the closest thing Queryable has to a twin, and on several counts it is the better deal. Ente is private, works offline and is open source — all three of the things Queryable is known for, at no cost.
- It is free. The app costs nothing; in gallery mode you tap “Continue without account” and keep everything on the phone, and a signed-in free plan includes 10 GB of encrypted storage.
- Its Magic Search “runs entirely on your device” and, after indexing, “works without an internet connection”.
- It is open source under AGPL-3.0, with 29,754 GitHub stars as of October 2026.
- It detects and groups faces on the device, and detects text in photos on the device so you can copy it.
- If you sign in, it adds end-to-end encrypted backup and sync, with indexes encrypted and stored on Ente’s server.
- It supports older systems: the US listing requires iOS 15.1 or later, against Queryable’s iOS 18.0.
Where Queryable adds something
Queryable is narrower than Ente — it is not a gallery and not a backup — and spends that focus on search and cleanup inside the Apple Photos library.
- Search with a sample photo, from the camera or the library, and a “Similar Photos” panel on every result. Ente’s pages read for this comparison do not document either.
- A three-tier cleanup from one scan — Exact Duplicates, Nearly Identical and Similar Scenes — with an aesthetic score choosing the keeper. Ente documents exact-hash deduplication plus a separate “Similar images” tool, and its gallery-mode feature table does not list that tool.
- Non-English queries are translated on the device and searched together with the original; Ente says English queries “typically work best”.
- A different model variant: Queryable runs MobileCLIP-S2, while Ente says Magic Search is based on MobileCLIP v1. That is a fact about the models, not a measured quality difference.
- One price, $4.99, with no storage plan to sell, and a Greater China edition, 寻隐, with a natively trained Chinese text encoder.
Compare
Every Ente cell is a fact Ente publishes — on ente.com, in Ente’s help centre, or in its US App Store listing — numbered to its source. Where Ente publishes nothing in those pages, the cell says “Not published”.
| Dimension | Ente Photos 1.3.66 | Queryable 2.1.6 |
|---|---|---|
| Price (as of October 2026) | Free to download, free with no account in gallery mode, and a free 10 GB plan. Paid storage on ente.com starts at $2.49/month or $29.88/year (Starter, 50 GB); US App Store in-app purchases start at 10 GB $0.99, period not stated on the listing. [1][2][3] | $4.99 on the US App Store, paid once. No subscription, no in-app purchases, no free tier and no trial. [10] |
| Account and backup | Optional. Gallery mode works “without creating an account” (since 13 March 2026); backup, sync, sharing and Ente albums need an account, and then indexes are encrypted and stored on Ente’s server. [3][4][5] | No account exists. None. Queryable reads the system photo library and never copies photos anywhere. The index lives on one device and is not synced or backed up; two devices means indexing twice. [11] |
| Requirements | iOS 15.1 or later (US listing, version 1.3.66). [2] | Any iPhone or iPad running iOS 18.0 or later. No Apple Intelligence requirement, no account and no sign-in. [10][11] |
| Search model | Magic Search “is based on Apple’s MobileCLIP model (v1)”. [5] | Apple’s MobileCLIP-S2 — not OpenAI’s CLIP — with 512-dimension embeddings, run through Core ML; search ranking runs on the Metal GPU. [11] |
| Where processing happens, and offline use | On the device: “Magic search runs entirely on your device … after that magic search works without an internet connection.” In gallery mode, “Processing happens locally and no data leaves your phone.” [6][3] | On the device, through Core ML and Metal; the app makes no network request of its own. Once a photo is indexed, searching for it works with networking switched off. Two boundaries: the one-time language-pack download for translated queries, and iCloud-only originals, which must be made available on the device before indexing. [11][12] |
| Query languages | Several languages are accepted; Ente says “English queries typically work best”. [6] | Type in your own language: the English edition translates a non-English query on the device with Apple’s Translation framework (a one-time free language-pack download) and searches with the original and the translation together. In mainland China, Hong Kong, Macao and Taiwan the app is sold as 寻隐 / 尋隱, with a natively trained Chinese text encoder. [11] |
| Search with an example photo | Not published in the Ente pages read for this comparison. [6] | Yes. A camera button in the search bar takes a photo or picks one from the library and uses it as the query, and every photo’s detail view has a “Similar Photos” panel. [11] |
| Faces | Yes, on the device, including in gallery mode. [3] | No face, people or pet recognition, and no albums, places or map. [11] |
| Text inside photos | On-device text detection, to copy text. Ente says Magic Search “is not a dedicated ‘search by detected text’ feature”. [7] | No. Queryable performs no text recognition of any kind; a screenshot is found by what it looks like, never by the words printed in it. [11] |
| Duplicates and similar photos | Exact duplicates by file hash, plus a separate on-device “Similar images” tool. Whether that tool works in gallery mode is not published. [8][3] | Three tiers from one 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 batch. Stills only: videos are never grouped or deleted. [11] |
| Open source | Yes: AGPL-3.0, 29,754 GitHub stars on 2026-10-30. [9] | Partly. The MIT-licensed core is public at mazzzystar/Queryable (more than 2,900 GitHub stars), but it stops at 29 March 2026: everything added in 2.1.2–2.1.6 is in a private repository. [13] |
| 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 in the project’s comparison research protocol, and this page is deliberately published without one.
Which to choose
Choose Ente Photos if…
- You want a free app, with nothing to buy at all.
- You want face grouping, copyable text, or encrypted backup and sync later.
- You want an AGPL open-source app, or your device runs an iOS version older than 18.
Choose Queryable if…
- You want to search with a sample photo or open similar photos from any result.
- You want near-duplicates and similar scenes sorted into tiers with the keeper picked for you.
- You search in a language other than English, or want the Chinese-native 寻隐 edition.
- You prefer one purchase to a free app with storage plans attached.
What this page does not claim
Ente is free and does more than Queryable in several rows; that is the point of the page.
- Neither app’s search quality was tested, and MobileCLIP-S2 versus v1 is not offered as proof that either finds photos better.
- Ente may support things its pages do not document — search by example image, or similar-image cleanup in gallery mode. Those cells say “Not published”, not “No”.
- Ente’s iOS in-app purchases are named by storage size only; their billing periods are not stated on the listing.
Methodology
Dates and storefront: Ente facts were read on 30 October 2026 from ente.com (pricing and machine-learning pages), five Ente help-centre articles, Ente’s 13 March 2026 release post, the GitHub repository and the US App Store listing for Ente Photos 1.3.66 (minimum iOS 15.1). Web prices are Ente’s own; iOS prices are the US in-app purchases as listed. Queryable is version 2.1.6, US App Store, minimum iOS 18.0.
What was tested: nothing. This is a facts-only comparison published under the Tier 1 rules of the project’s comparison research protocol: every Ente cell is a fact Ente publishes, with its source numbered; every Queryable cell describes the shipping app; anything neither side publishes says “Not published”, and result quality is marked “Not tested”. No row states a ranking, accuracy, recall or speed figure. Queryable publishes this page, so read it as our argument checked against Ente’s own pages, not as a neutral review. Corrections are welcome through the project’s GitHub repository.
Citations
Every Ente Photos statement above was read from the pages below on the date shown. Vendor pages change without notice; if one now says something different, the vendor is right and this page is stale.
- Ente — Pricing
- Ente Photos — US App Store listing (v1.3.66)
- Ente Help — Gallery mode
- Ente blog — New in Ente Photos: Offline Gallery… (13 March 2026)
- Ente — Machine learning (technology page)
- Ente Help — Magic Search
- Ente Help — Detect Text (OCR)
- Ente Help — Storage optimization
- GitHub API — repos/ente-io/ente
- Queryable on the US App Store Source for the one-time price, the absence of in-app purchases, the minimum iOS version and the “Data Not Collected” privacy label.
- 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.
- Offline AI photo search The iCloud-only boundary and what “offline” does and does not cover during indexing.
- 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 (translation, search by photo, time filters, three-tier cleanup).
Questions about Ente Photos and Queryable
Is Ente Photos free?
Yes. The app is free, it can be used as a gallery with no account, and the signed-in free plan includes 10 GB of storage. Paid storage starts at $2.49 a month for 50 GB on ente.com (as of October 2026). Queryable costs $4.99 once.
Does Ente Photos search on the device?
Yes. Ente says Magic Search runs entirely on your device and works without an internet connection after indexing, and that it is based on Apple’s MobileCLIP model (v1). Queryable also searches on the device, with MobileCLIP-S2.
Is Queryable the only open-source offline photo search for iPhone?
No. Ente Photos is open source under AGPL-3.0 and searches on the device. Queryable’s MIT-licensed core is public, but the public repository stops at 29 March 2026 and the features added since are developed privately.
What does Queryable do that Ente does not document?
Search with a sample photo, a “Similar Photos” panel on every result, a three-tier cleanup (exact, nearly identical, similar scenes) with an aesthetic keeper, and on-device translation of non-English queries. Ente’s pages read for this comparison do not document the first two.
Can I use Ente and Queryable together?
Yes. Both read the iPhone’s photo library. You could keep Ente for faces and encrypted backup and use Queryable for sample-photo search and similar-photo cleanup.