Frequently Asked Questions

Quietra was built for two overlapping groups:

Privacy-conscious professionals — lawyers reviewing case notes and contracts, doctors handling patient-adjacent information, accountants working with financial records, and HR professionals managing confidential files. For these users, uploading documents to ChatGPT or any cloud AI is categorically off the table — not just for privacy preference, but often for regulatory or ethical reasons.

Privacy-motivated individuals — people who want an AI assistant for personal journaling, health tracking, financial planning, or sensitive reflection without their most private thoughts being stored on and processed by a third-party server.

Quietra is not a developer tool. It's a polished app designed to "just work" for non-technical users.

Documents with clear, structured text work best:

• Contracts and legal agreements — NDAs, employment agreements, service contracts
• Reports and briefings — financial reports, research summaries, policy documents
• Medical records and clinical notes — discharge summaries, lab reports, referral letters
• Academic papers — literature reviews, study results, technical documentation
• Business documents — meeting notes, proposals, project plans

Documents that are mostly images (scanned PDFs without embedded text), tables of raw numbers, or highly specialized technical notation (mathematical formulas, code) may yield lower quality responses.

They serve fundamentally different needs:

Feature	Quietra	ChatGPT
Data sent to cloud	Never	Always
Works offline	Yes	No
Document Q&A with citations	Yes	Partial (paid)
Document-aware AI	Yes	No
Open-ended world knowledge	Limited	Extensive
Subscription required	No	Yes (for GPT-4)
Suitable for confidential docs	Yes	No

Bottom line: Choose Quietra when privacy matters and you're working with your own documents. Choose ChatGPT when you need broad world knowledge, code generation, or complex reasoning at scale — and privacy is not a concern.

For document-grounded tasks — summarization, Key Info, comparison, translation, and Q&A from provided text — Quietra performs very well. Its proprietary document intelligence models understand what your document is and tailor processing accordingly, working alongside an on-device LLM to deliver precision that approaches cloud AI for structured documents like contracts, medical records, and financial forms.

For open-ended general knowledge, complex multi-step reasoning, creative writing at scale, or up-to-date information about the world, cloud models like GPT-4 have a significant advantage.

Quietra is not trying to replace ChatGPT. It fills a different need: private, document-aware AI that is architecturally impossible to offer on cloud-first tools.

Absolutely — they complement each other. A common workflow: use Quietra to review a confidential document and get the Key Info or a summary (keeping it private), then use that higher-level write-up — not the raw document — in ChatGPT for further analysis or drafting. This way, sensitive source material never reaches the cloud, but you still benefit from larger model capabilities for synthesis and writing.

Quietra combines proprietary document intelligence models with an on-device LLM — all running entirely on your device.

When you import a document, Quietra's proprietary models identify what kind of document it is — a lease, an NDA, a medical record, a tax form, and dozens more. They tailor every operation to that specific document type: the right information to surface, the right details to summarize, the right terms to look for. The on-device LLM then generates responses grounded in your document.

All models are developed by Quietra, built into the app, and run on your device. They carry the same privacy guarantee as everything else in Quietra: zero cloud dependency, zero data collection.

Most AI apps send your document to a cloud server and run it through a single general-purpose model that treats every document the same way.

Quietra is different in two ways. First, nothing leaves your device — ever. Second, Quietra's proprietary models understand what your document is before the LLM processes it. They know that a lease needs rent schedules and security deposits, while an NDA needs confidential information definitions and survival periods. This document-aware approach, combined with the on-device LLM, delivers more precise, more relevant results than a generic cloud model.

Quietra recognizes dozens of document types, including:

• Legal agreements — NDAs, employment agreements, leases, service agreements, licenses, loans, and more
• Medical documents — lab reports, prescriptions, patient records, insurance claims
• Financial documents — invoices, purchase orders, expense reports, tax forms, financial statements
• Government forms — applications, permits, regulatory filings

Each type gets its own specialized document signals and AI instructions, so Quietra can surface the most relevant information quickly. Documents that don't match a specific type still work well with Quietra's general-purpose analysis.

We believe in transparency about what Quietra can and cannot do well:

Model capability ceiling: The on-device model is smaller than large cloud models. It may struggle with very long or complex reasoning chains, nuanced legal interpretation, or specialized technical domains.

Hallucinations: Like all language models, Quietra can generate plausible-sounding but incorrect information. For document Q&A, always check the cited passages in the source document. Never rely solely on Quietra output for consequential decisions.

No real-time information: The model has a training knowledge cutoff and no access to live web data (unless you enable online mode). It cannot answer questions about recent events.

Scanned documents: Very low-quality scans, handwritten text, or documents with complex layouts (multi-column, dense tables) may not OCR cleanly, leading to lower-quality answers.

Language support: The model performs best in English. Other languages may work but with reduced accuracy.

No professional advice: Quietra's outputs are not legal advice, medical advice, or financial advice. Always consult a qualified professional for decisions with real consequences.

Yes — Quietra can and does make mistakes. AI language models are probabilistic systems, not deterministic fact-checkers. They can:

• Misread or misinterpret a passage in your document
• Generate a plausible but incorrect summary
• Miss important nuances, exceptions, or caveats in legal or medical text
• Confuse similar clauses or sections

To mitigate this, Quietra always shows you the source passages it used for document-grounded answers. Use these citations to verify the AI's reasoning against the original text. For any decision with real-world consequences — signing a contract, making a medical decision, acting on financial analysis — always verify with the source document and consult a qualified professional.

Quietra can process documents and respond in multiple languages, but English is where the on-device model performs most reliably. European languages (French, German, Spanish, Italian, Portuguese) generally work well. Results may be less consistent for less-represented languages. If accuracy is critical for a non-English document, always verify outputs against the source text.

Yes — this is an important limitation to understand. While Quietra itself stores all data within the app sandbox, the operating system and other device-level mechanisms may independently retain traces of your activity. Examples:

OS-level memory: Content from Quietra (like any app) passes through device RAM during processing. While RAM is cleared when the device is restarted or memory pressure forces a clear, it is theoretically accessible to the OS or to apps with elevated system privileges.

Screenshots and screen recording: If you take a screenshot of a Quietra conversation, that image is saved to your camera roll — outside Quietra's sandbox — and may be backed up to iCloud or synced to other devices.

Accessibility services: If you have screen readers, accessibility tools, or third-party keyboard apps with full access enabled, they may read on-screen content from Quietra. Review which accessibility services have access on your device.

iCloud Backup: By default, iOS may back up the Quietra app sandbox (including your conversations and documents) to iCloud, encrypted under your Apple ID. This backup is not visible to us, but it does exist on Apple's servers under your account.

Quietra cannot prevent these OS-level behaviors — they are outside the app's control. What Quietra can guarantee is that it itself never transmits your data anywhere.

Quietra eliminates the most significant data leakage vector in AI tools — sending your content to a remote server for processing. However, residual risk exists at the device level that you should be aware of:

Risks Quietra prevents:
• Cloud AI server receiving your documents or prompts
• Third-party company storing your conversation history
• Data breaches of a company's AI infrastructure exposing your content

Residual risks outside Quietra's control:
• Device theft or physical access: If someone gains physical access to your unlocked device, they can read content on screen or within the app. Use a strong passcode and Face ID/Touch ID.
• iCloud Backup: Your device backup (including Quietra data) may be stored on iCloud. Secure your Apple ID with strong credentials and two-factor authentication.
• Jailbroken devices: On a compromised device, OS-level isolation protections do not apply. Quietra should not be used for sensitive documents on a jailbroken device.
• Third-party keyboards: Keyboards with "full access" can read what you type. Use the system keyboard when entering sensitive content.
• MDM/enterprise device management: On a company-managed device, your IT administrator may have elevated access capabilities. Do not use a work-managed device for personal private content.
• Screenshots: Any screenshot of a Quietra screen escapes the app sandbox.

No software can provide absolute privacy guarantees once device-level security is compromised. Quietra's approach minimizes risk by design — it does not create a cloud-side attack surface — but device security remains your responsibility.

Possibly — this depends on your iOS backup settings. iOS may include Quietra's app sandbox in your iCloud Backup by default. This backup is:

• Encrypted and stored under your Apple ID
• Not accessible to us (we never receive it)
• Subject to Apple's terms and privacy policy

If you want to ensure Quietra data is excluded from iCloud Backup, go to: Settings → [Your Name] → iCloud → Manage Account Storage → Backups → [Your device], and turn off Quietra from the list of apps being backed up.

Alternatively, if you use Advanced Data Protection in iCloud, your backup is end-to-end encrypted and even Apple cannot access its contents.

This depends on your organization's MDM (Mobile Device Management) policy. On a corporate-managed device:

• Your IT administrator may have capabilities to remotely wipe, monitor app usage, or in some configurations access data on the device
• Quietra's privacy architecture prevents us from receiving your data, but it cannot prevent a device-level MDM with elevated privileges from doing so

For personal sensitive use (journaling, private documents), use Quietra on your personal device rather than a corporate one. For organizational use of Quietra (e.g., reviewing company contracts), your IT team should assess compatibility with your MDM policy.

Most "private" AI apps rely on privacy policies — they promise not to use your data, but they do receive it on their servers. Quietra is different by architecture: AI inference happens entirely on your device, with no cloud component. Proprietary models and the on-device LLM run directly on your device. The only network calls Quietra ever makes are user-initiated and unrelated to AI processing — for example, license activation and the contact form. Your documents, conversations, and AI prompts never leave your device.

No. Quietra contains no analytics SDK, no crash reporting to external services, and no usage tracking of any kind. We do not know how you use the app, how many documents you've imported, or what you ask the AI. Your interactions are not a product and are not monetized.

Quietra is designed specifically for this use case. Lawyers, doctors, accountants, and financial professionals regularly use Quietra to work with documents that cannot be uploaded to cloud AI tools. Because everything stays on your device, there is no data transmission that would create HIPAA, attorney-client privilege, or data sovereignty concerns. However, Quietra is a productivity tool — not a regulated compliance product — and you remain responsible for your own compliance obligations. Consult your compliance officer or legal counsel if you have specific regulatory requirements.

No. Once installed, Quietra works completely offline. All features — document import, AI chat, Q&A, summarization — work without a Wi-Fi or cellular connection. You can use Quietra in airplane mode.

Quietra requires a device capable of running on-device AI models. This currently means:

• iPhone 15 Pro and iPhone 15 Pro Max
• iPhone 16, 16 Plus, 16 Pro, and 16 Pro Max
• Newer iPhone models as they are released

Standard iPhone 15 and 15 Plus are not supported because they lack on-device AI capability. If your device is incompatible, Quietra will display a clear explanation and guidance.

Quietra requires iOS 26.2 or later. This is required for the on-device LLM framework that Quietra uses for AI inference. You can update by going to Settings → General → Software Update.

No. The on-device LLM is provided by iOS — it's built into the operating system on supported devices. Quietra's proprietary document intelligence models are built into the app itself. There is nothing extra to download or configure. If the on-device LLM is not yet available, iOS will typically download it in the background when Apple Intelligence is enabled in Settings.

Quietra will display a compatibility screen explaining that on-device AI is not available on your device or iOS version. There is no cloud fallback — we do not silently switch to a remote model. This is a deliberate design choice: Quietra's privacy guarantee only holds when all processing runs locally. The screen will provide guidance on how to enable on-device AI or upgrade your device.

At launch, Quietra will be available for iPhone only (iOS 26.2+). A macOS version for Apple Silicon Macs is planned for Q3–Q4 2026. An iPad version may follow alongside the macOS release. No Android or Windows version is planned.

Quietra supports:

• PDF — text extraction with support for scanned (image-based) PDFs via on-device OCR.
• TXT and MD — plain text and Markdown files.
• DOCX / DOC — Microsoft Word documents.
• RTF — Rich Text Format files.
• JPG, PNG, HEIC — photos and screenshots, with on-device OCR text extraction.

All OCR and text extraction happens locally. No document content is sent to any external service.

When you import a document, Quietra processes and indexes the text on your device. When you ask a question, the app retrieves the most relevant sections and passes them to the on-device AI model as context. The model then generates an answer grounded in those passages, and Quietra shows you which sections of the document were used — so the AI is answering from your document, not from its general training knowledge.

There is no hard-coded file size limit. However, very large documents (hundreds of pages) will take longer to index and may use more device storage for the local search index. Performance depends on your device — newer iPhones with more RAM will handle large documents faster. For best results with extremely large documents, consider splitting them into smaller logical sections before importing.

Yes. Quietra uses the iOS Files picker, which gives you access to any file location your device can reach — including iCloud Drive, Dropbox, Google Drive (via their iOS apps), and local device storage. Once you select and import a file, Quietra processes it entirely on-device. The original file stays in its cloud location; only the extracted text is stored locally in Quietra's sandbox.

Quietra's proprietary models combined with the on-device LLM are optimized for document tasks: summarization, Key Info, comparison, classification, translation, and contextual Q&A. For these use cases they perform well. However, like all AI models, it can make mistakes — it may occasionally misread a passage, miss nuance, or hallucinate details. Always review AI-generated summaries or Key Info against the source document for any decision with real consequences. Quietra shows you the source passages used for each answer so you can verify the reasoning.

Yes — Quietra has a general chat mode that can answer everyday questions, help with writing, brainstorming, and reasoning. However, the on-device LLM has limitations compared to large cloud models for open-ended factual questions or complex reasoning. It excels at document-grounded tasks. For general knowledge questions, treat its answers as a starting point and verify when accuracy matters.

Yes. In the Settings tab, you can delete all local data with a single tap — conversations, imported documents, and the document search index. You can also delete individual documents from the Documents list. Deleting the app from your iPhone removes all data entirely, as everything lives in the iOS app sandbox.

Quietra stores all data in the iOS app sandbox. Whether that sandbox is included in iCloud Backup depends on your iPhone's iCloud Backup settings, which is controlled by iOS — not by Quietra. If you back up your iPhone to iCloud, the Quietra app data may be included in that backup. This backup is encrypted and stored under your Apple ID, not accessible by us. If you want to ensure data is never backed up, you can disable iCloud Backup for Quietra in Settings → [Your Name] → iCloud → Manage Storage → Backups.

Quietra has three tiers. Free — unlimited AI chat and 5 documents per month (every feature unlimited on each document), forever, no credit card required. Pro ($49.99) — a one-time purchase that unlocks unlimited documents for personal use. No subscription, no recurring cost. Business ($9.99/month or $99.99/year) — everything in Pro, plus authorization for commercial and organizational use. Cancel anytime. The current public Business offering is a single-seat commercial entitlement. If we later launch multi-seat admin or volume deployment tooling, we'll describe that separately. We do not monetize your data — all AI processing is on your device.

Quietra launches in an Early Access period while we iterate quickly based on real usage. During Early Access, the Free tier's monthly document limit is suspended — Free users can analyze an unlimited number of documents. The Early Access window has a defined end date disclosed inside the app. After Early Access ends, the Free tier returns to 5 documents per month. Pro and Business users are unaffected either way.

Yes — permanently. Pro licenses purchased during Early Access entitle you to unlimited document analyses in perpetuity, even after the Free tier's 5-per-month cap returns. This is memorialized in our Terms of Use, Section 3.4. If you're considering Pro, buying during Early Access is the lowest-risk moment to do so.

Use the contact form, or the in-app Settings → Report a Bug screen. Both go to the same inbox. Early Access feedback directly shapes what we ship — please err on the side of writing in.

Because Quietra runs entirely on your device. There are no cloud servers, no API costs, and no ongoing infrastructure expenses per user. A subscription would charge you for services we don't provide. Pay once, use forever.

Personal use means using Quietra for your own private, non-commercial purposes — personal journaling, reviewing your own documents, learning, creative writing, and similar activities. The Free and Pro tiers are for personal use. Business use means using Quietra as part of your work at a company, law firm, medical practice, financial institution, or any other organization. Business use requires a Business license.

At launch, you'll be able to purchase in the pricing section on our homepage (via Stripe, US only) or directly in the app (via the App Store, worldwide). Purchasing on the website has lower fees, which helps support Quietra's development. Your license will sync across devices via your email address.

After launch: open Quietra on your new iPhone, go to Settings → License → Activate Purchase, and enter the email address you used at checkout. We'll send a 6-digit verification code to that email — enter it to activate. Your license syncs across platforms as more become available (Android and Mac planned). You'll also be able to manage your license on the web at quietra.ai/account.

Unlimited devices under your Apple ID (all your iPhones and iPads signed into the same Apple ID count as one slot), plus 2 additional cross-platform slots for web activations or future Android devices. If you hit the limit, you can deactivate a device from Settings or the account page. No device transfer hassle — we keep it simple.

Email us at info@quietra.ai with proof of purchase (the Stripe or App Store receipt). We'll verify ownership and transfer your license to a new email address. Because we don't use passwords, there's nothing to reset — just a simple email change.

Business subscribers can cancel at any time. App Store purchasers manage their subscription in Settings → Apple ID → Subscriptions on iPhone. Website purchasers should contact info@quietra.ai for cancellation help until self-service billing tools are live. In all cases, you'll keep access until the end of the current billing period.

When you delete Quietra from your iPhone, iOS removes the app sandbox entirely. All conversations, imported documents, and the document search index are permanently deleted. If you created a license account for cross-platform sync, a cryptographic hash of your email and your license status remain on our servers. You can request deletion through the Contact form or the in-app license settings.

All Quietra data is stored in the iOS app sandbox under Application Support/QuietraStorage/. This directory contains three subdirectories: conversations/ (chat history), documents/ (imported file metadata and extracted text), and index/ (the local document search index). This data is isolated by iOS and inaccessible to other apps.

Yes, via the standard iPhone transfer mechanisms. If you use "Transfer from iPhone" or restore from an iCloud or encrypted iTunes backup when setting up your new device, Quietra and its data will be transferred. Direct device-to-device transfers (Quick Start) also transfer app data. Manual export/import of individual documents is not yet supported but is planned for a future release.

Quietra's current public Business offering covers commercial use for a single user. Organization-managed MDM / Apple Business Manager / volume deployment support is not live yet. If you're evaluating Quietra for a larger team or managed-device rollout, contact us at info@quietra.ai and we can tell you the current status.

Quietra's architecture — no data transmission, no cloud processing, no external services — is well-aligned with the data minimization principles of GDPR and the security requirements of HIPAA. However, Quietra is a consumer productivity application, not a certified compliance product. We do not offer a Business Associate Agreement (BAA) at this time. For regulated healthcare or financial environments, consult your compliance officer before deployment. We are happy to provide technical documentation about our data flows (or absence thereof) to support your compliance review — contact us through the Contact form in the app.

Yes. A macOS version for Apple Silicon (M-series) Macs is planned for Q3–Q4 2026. Quietra is built with Swift and SwiftUI, which means the iOS and macOS codebases share a significant portion of their logic, making the macOS port considerably faster to build than starting from scratch. The macOS version will target the same on-device AI approach.

Not in the foreseeable future. Quietra's architecture requires a capable on-device LLM provided by the operating system, which is currently available only on iOS and macOS. Android and Windows don't have an equivalent, which would require a fundamentally different technical approach. We're focused on the Apple ecosystem where the technology genuinely enables our privacy goals.

Planned for v1.1 and beyond:

• Camera capture — point your camera at a physical document and Quietra OCRs it on-device
• Private journaling — a dedicated journaling workspace with AI reflection features
• Enhanced DOCX/RTF import — formatting-aware text extraction
• Data export — export your conversations and documents to local files

Recently shipped:
• Document comparison — compare two versions of any document, even across formats (PDF vs scan vs Word). Quietra detects every change, classifies them by significance, and presents a structured change report with filtering.

We ship updates when features are ready and meet our quality bar — not on a fixed schedule.

Use our contact form with a description of the bug or feature request. For bugs, include your iPhone model, iOS version, and steps to reproduce.