Offline PDF Editor for Android

Why offline matters for documents

On a computer or without installing an app, the free offline PDF tools can merge, split, extract, reorder, rotate, sign and convert images in your browser. Load the chosen tool while connected, then test a sample offline. They do not edit existing text or run OCR. For those tasks, see the Android app download page.

The files people edit on a phone are the ones they least want copied: contracts, ID scans, medical results, invoices, HR paperwork. Every cloud-based editor requires an upload, and an upload is a second copy of the document on someone else’s server — subject to retention policies, breach exposure, and terms that can change. An on-device editor has no server to breach, leak, or hand over.

Local document processing and network access are separate questions. The Android app declares INTERNET permission and includes Firebase Analytics for optional usage statistics. That does not turn editing into a cloud service: documents, file names, file locations and search terms are excluded from usage events. The app privacy policy describes the data fields and controls.

Document work can continue in airplane mode, on metered connections and in areas without coverage. Documents are not queued for upload. If you opt into usage statistics, pending usage events may be stored locally for up to seven days; this is separate from your PDFs.

What runs on the device

The document operations below run on the phone, and none of them uploads a file. The app is on Google Play and in early access, so every feature is free during this period. See the app terms for how purchases will work and the download page for the store link.

• Reading, search and the document diagnosis.
• Annotation: highlight, underline, strikethrough, ink, notes.
• Page tools: delete, reorder, rotate, duplicate, insert, crop, flatten, repair.
• Text editing of existing content, with text shaping and font subsetting on the phone.
• Find and replace across the document, and text comparison between two files.
• OCR, with the recognition model bundled in the APK and on-device printed-or-handwritten line classification.
• PDF to Word, Excel and PowerPoint through a self-contained OOXML writer.
• Password protection and removal; manual and rules-assisted redaction; watermark and metadata tools.
• Merge, split, header & footer and compression; verified saving with a change report; undo and crash-safe drafts.

The conversion writer is worth a note: DOCX, XLSX and PPTX files are produced by a PDFMason-built OOXML writer running on the phone. No conversion service, no server-side office suite, no queue. Heavy jobs show progress, run in the background, and can be cancelled — the trade for offline is compute, not capability.

The app’s boundaries are stated in full on the PDF editor for Android page. The short version: line-and-arrow annotation shapes were cut after real-document testing, and text editing refuses Indic scripts and vertical text rather than writing them wrong. Form filling — text fields, checkboxes and radio buttons — works and survives saving; non-Latin field values are embedded with a shaped font subset at save. None of those limits exist because of the offline design — they are honesty about a first version, not network workarounds.

Works without Google Play services

Document processing does not depend on Google services. The OCR engine is ML Kit in bundled form: the recognition model ships inside the APK instead of being downloaded from Google Play services. A phone with no Google Mobile Services — Huawei and HarmonyOS devices, de-Googled ROMs, locked-down enterprise handsets — gets the same editing, OCR, conversion and password tools as a Pixel.

Camera scanning is the one feature that is different. It uses Google’s document scanner, so the scan entry appears only on devices that have Google Play services, a camera and enough free memory; on other devices it stays hidden. The photo-import path — turning pictures already in your gallery into a PDF — has no such requirement. And a camera scan only produces a picture PDF: giving it a searchable text layer is the bundled OCR’s job, which runs on any device.

Bundling forced removals, too. The bundled recognizer normally carries a telemetry backend behind it, plus permissions for the advertising ID and network state. Those declarations are stripped from the manifest, because an offline document tool has no business holding an advertising ID. The recognizer works without any of them.

Google services are not used to process your documents. If you enable usage statistics, Google processes those events through Firebase Analytics. Its pseudonymous installation identifier and technical device information mean this data is not fully anonymous.

Documents and optional usage statistics

No account or sign-up is required. Usage statistics start only after an explicit choice at first use or in Settings → Privacy. The app does not collect an advertising ID. Crash diagnostics are unavailable in this version: the Crashlytics SDK is not included and crash reports are not sent.

Files stay where you put them. The app reads what you pick in the system picker and writes where you tell it to save; it does not request all-files access.

The website and Android app have separate privacy policies. The website uses a first-party usage beacon without document contents or file names, and the download page has an optional form for iPhone-version news that asks for an email address. It loads no third-party analytics scripts or font CDN. See the website privacy policy for the fields it records.

To check local processing after installing the app, save a test PDF on your phone, switch to airplane mode, edit it, save a copy and reopen it. Separately review Settings → Privacy for your statistics choice. For the browser tools, inspect the Network panel while processing a sample: document data should not appear in requests. Page assets and first-party usage events are distinct from file uploads.

FAQ