# WhattaRAG — Chat with your docs, privately

> Chat with your docs, privately (iPhone/iPad app by Fernando Putallaz.)

- Source: https://appshunter.io/ios/app/whattarag/id6767934110 (this page in markdown: same URL + `.md`)
- Developer: [Fernando Putallaz](https://appshunter.io/developer/1557728137)
- Category: Productivity, Reference
- Price: Free
- Age rating: 4+
- Requires: iOS 26.0 · 4 MB
- Languages: American English
- Released: 2026-06-10
- Data updated: 2026-08-26
- User reviews in markdown: https://appshunter.io/ios/app/whattarag/id6767934110/reviews.md

## What is WhattaRAG?

WhattaRAG lets you ask questions about your own documents and get answers drawn from them — without anything leaving your iPhone.

Import or paste your text and markdown documents, then start a conversation. WhattaRAG finds the passages that matter and uses Apple Intelligence to answer in plain language, with tappable citations back to the exact source.

Everything happens on-device. No accounts, no sign-in, no servers. Your documents, conversations, and everything the app generates stay in a private database on your Phone. 
The app makes no network requests of any kind — there's nothing to opt out of, because there's nothing being collected.

How it works
  • Add a document by importing a text or markdown file, or by pasting text directly.
  • Start a conversation scoped to one document, or chat without a document.
  • Ask a question. WhattaRAG retrieves the relevant passages and answers from them.
  • Tap a citation to jump straight to the source passage.

  Private by construction
  • All AI runs on-device using Apple Intelligence and Apple's Natural Language framework.
  • No analytics, no crash reporters, no advertising SDKs, no telemetry.
  • No accounts and no identifiers — nothing connects the app to you, because nothing reaches us.
  • Clear everything from Settings at any time, or remove the app to erase its data completely.
  
Honest about the limits
WhattaRAG answers from the most relevant excerpts of your documents, not your entire library at once. It's excellent at finding and explaining specific passages. It's not built to count, rank, or summarize a whole collection in a single answer — and it tells you when a question falls outside what retrieval can do well.

Requirements
WhattaRAG needs an iPhone with Apple Intelligence running iOS 26 or later (iPhone 15 Pro and newer). The on-device language model is provided by iOS; if your device or region doesn't support Apple Intelligence yet, the app will tell you.

v1 works with text and markdown documents (more formats coming next). Built by a solo iOS developer who cares about the details.


## Version history (last 4 releases)

### 1.6.0 — 2026-08-19

Retrieval and answer quality improvements throughout: more accurate answers on longer documents, more available context in conversations, and faster response times. Several retrieval accuracy fixes as well. Your library and conversations carry over as always.

### 1.5.0 — 2026-07-29

This one's all under the hood: a rework of how your documents and conversations are stored, making WhattaRAG more reliable and laying the groundwork for what's next. Your existing library and chats carry over automatically.

### 1.0.1 — 2026-06-22

This release improves answer quality across the board.

• Smarter search. WhattaRAG now searches your documents two ways at once — by meaning and by exact wording — so it finds the right passage more often.

• Forgiving questions. Vague or misspelled questions are tidied up before searching, so you get good answers without phrasing things perfectly.

• Sharper results in long documents. Each passage now carries its section heading into the search, which improves accuracy in larger files.

As always, everything runs entirely on-device. Nothing you import or ask ever leaves your iPhone.

### 1.0 — 2026-06-11

No release notes.

## More apps by Fernando Putallaz

- [Mindstorm](https://appshunter.io/ios/app/mindstorm/id1600252610)

All apps by Fernando Putallaz: https://appshunter.io/developer/1557728137

---

*Data collected daily from the US App Store and indexed by [AppsHunter](https://appshunter.io/). User reviews are verbatim App Store reviews. Ratings, prices and chart positions refresh continuously; this snapshot is from 2026-08-26.*