What is Lexicon?

Lexicon is a local-first desktop writing assistant that merges a distraction-free rich-text editor with built-in proofreading and AI-powered rewriting. It runs on Windows, macOS, and Linux without requiring an account, subscription, or cloud connection. The editor handles headings, lists, tables, code blocks, LaTeX math, slash commands, and automatic local saving. Proofreading relies on an embedded LanguageTool engine for grammar, spelling, and style checks with inline suggestions. AI capabilities such as rewrite, tone adjustments, summaries, and structural transforms run locally via bundled llama.cpp with quantized GGUF models in three size tiers, or can connect to a personal Ollama or LM Studio server. Documents can be imported from plain text, Markdown, HTML, and Word formats; exported to several output formats including PDF, EPUB, styled HTML, and Word with tracked changes. The app includes a personal dictionary, fully customizable keyboard shortcuts, multiple themes and paper textures, and a beta browser add-on for Chrome and Firefox that links to the running desktop app. All processing remains on the device by default; the network is only used for optional model downloads, update checks, or user-configured remote endpoints.

Should you self-host Lexicon?

Lexicon scores 47/100 on selfhost.directory (how scoring works) and ranks #47 of 81 in Notes & Editors. It has 222 GitHub stars.

It shipped 3 stable releases in the past year. The latest, v0.10.1, came out on 12 August 2026. New versions land roughly every 2 days.

Setup is easy: most people get it running in one sitting. It is MIT-licensed and written mostly in JavaScript.

Its closest alternative on selfhost.directory is Sovereign AI Workstation (48/100, 108 stars). It scores higher overall, so compare the two before you pick. See Lexicon vs Sovereign AI Workstation side by side.

Features

Auto-update checker
Browser ExtensionBeta; Firefox unsigned (temporary install only); Google Docs not supported; desktop app must be running.
Customizable keyboard shortcuts
Deep Proofread (hybrid)
Express in English (multilingual transcreation)
Focus/distraction-free mode
GPU Acceleration
Import/export multiple formats
LaTeX math support
Local AI rewrites & tone shifts
Multiple themes & paper textures
Offline grammar & spell checking
Personal dictionary
Privacy-first local processing
Rich Text Editing

How to install Lexicon

Install Lexicon with Linux

DEB (.deb) - Debian, Ubuntu, Linux Mint, Pop!_OS, etc.

To install the .deb package on Ubuntu, Debian, Linux Mint, or Pop!_OS:

Double-click the downloaded .deb file to install via Software Center.
Or run sudo dpkg -i Lexicon_*.deb (or sudo apt install ./Lexicon_*.deb) in Terminal.

Also available as .AppImage (per release workflow).

Install Lexicon with Windows

x64 - most modern Intel and AMD PCs
ARM64 - Windows ARM devices
x86 - older 32-bit Windows systems

The Windows installer bundles the app backend and its runtime. You do not need Python, Node.js, or Java to use the installed application.

Windows builds are not yet Authenticode code-signed. That can trigger SmartScreen and occasional antivirus false positives (Windows Defender, AVG, and similar). Lexicon is open source and MIT-licensed; only download installers from this repository's Releases page.

**SmartScreen ("Windows protected your PC"):

Click More info
Click Run anyway

Antivirus quarantine / "malicious code" warnings: these are usually reputation heuristics on new or unsigned indie installers, not a confirmed infection. If Defender or another AV blocks the file, restore/allow it from quarantine after you have confirmed the download came from the GitHub Releases URL above.

Not code-signed; SmartScreen and antivirus warnings expected.

Install Lexicon with macOS

Apple Silicon / arm64 - M-series Macs
Intel / x64 - Intel Macs

The macOS build is currently pre-release (unsigned). Gatekeeper may block launch or prevent the background engine from spawning because the app is not notarized by Apple. That is an unsigned-developer warning, not a claim that the build contains malware. Only download builds from this repository's releases page.

If Gatekeeper blocks Lexicon:

GUI: Control-click (right-click) Lexicon.app → Open → Open
System Settings: Privacy & Security → Open Anyway (Apple's guide)
Terminal: xattr -cr /Applications/Lexicon.app to clear the download quarantine flag

Unsigned/not notarized; Gatekeeper warnings expected.

FAQ

Can I change Lexicon's keyboard shortcuts?

Yes. The Shortcuts panel in Settings lets you remap proofread, suggestion, formatting, and focus commands with conflict detection. In the browser extension, Alt+Shift+L proofreads the focused field.

Do I need a GPU to use Lexicon AI?

No. Lexicon runs on CPU alone; a graphics card simply speeds up local AI. Light fits ~2 GB VRAM, Standard ~6 GB, Quality ~18 GB.

Can Lexicon translate my writing into English?

Yes. Express in English takes a short thought in Chinese, French, German, Hindi, Spanish, or dozens of other languages and phrases it as natural English in six tones, starting with Auto to keep the source register.

Why use Lexicon instead of a cloud grammar checker?

Cloud checkers bill monthly and process documents on their servers. Lexicon is a free desktop app with offline grammar and local AI options; network is only used for updates, model downloads, or servers you configure.

Why does Windows or macOS warn that Lexicon is unsigned or unsafe?

Desktop builds are not yet code-signed (Windows) or notarized (macOS). SmartScreen, Gatekeeper, and some antivirus tools warn on new or unsigned indie installers — this is a reputation/signing gap, not proof of malware.

Is there a browser extension?

Yes, a beta extension for Chrome and Firefox proofreads visible text fields through the desktop app. It requires the desktop app to be running and does not support Google Docs yet.

Can I proofread without an internet connection?

Yes. The packaged LanguageTool engine handles spelling, grammar, punctuation, and style locally when you run a check, so proofreading works offline.

Do I need a Lex model just to proofread?

No. Grammar and style checks work without an AI model. Download a local model only when you want Lex rewrites, or connect your own Ollama/LM Studio server.

Does Lexicon upload my writing to the cloud?

Not to a Lexicon-operated cloud service. Built-in proofreading and local AI run on your machine; only a remote endpoint you configure may receive text sent to it.