LeanBitLab/LeanType

LeanType is an Open-source keyboard, fork of HeliBoard with additional features and customisation

LeanType – A Privacy‑First, AI‑Enabled Android Keyboard

What it is – LeanType is an open‑source Android input method (IME) that builds on the lightweight, privacy‑focused HeliBoard codebase. It adds a long list of modern productivity tools, most of which can run offline on the device, and optional cloud‑based AI services for proofreading, translation, and text generation.

Key capabilities

  • AI proofreading & generation – Connect to Google Gemini, Groq, OpenAI, or any self‑hosted OpenAI‑compatible endpoint (Ollama, LM Studio, LocalAI, etc.). The offline flavor can also run quantised GGUF models locally via llama.cpp for on‑device grammar checking.
  • Voice typing – On‑device Whisper speech‑to‑text through a separate Voice Plugin.
  • Handwriting & OCR – In‑keyboard camera viewfinder and screenshot OCR powered by ML Kit, also provided via plugins.
  • Translation – Offline plugin or cloud AI, selectable per‑language.
  • Math evaluation – Inline calculation of expressions typed after =.
  • Rich customisation – Sound packs, key‑shape/theme sliders, per‑app profiles, custom toolbar keys, and up to five saved layout profiles.
  • Privacy‑first design – No sensitive permissions (no contacts, no system‑alert‑window), optional internet access only for cloud AI or update checks.

Flavors

Flavor When to use AI support
Standard (recommended) General users, F‑Droid or direct APK install Cloud AI + optional offline GGUF via plugin
Offline Air‑gapped or maximum‑privacy devices No internet; runs only local GGUF models

Both flavors are under 11 MB and work on Android 6.0+ (standard) or Android 5.0+ (offline).

How it works

  1. Install the base APK (standard or offline).
  2. Add desired plugins (Voice, OCR, Translation, Handwriting, Offline‑AI) from the linked GitHub releases.
  3. For cloud AI, supply an API key and model ID; for offline AI, load a GGUF model file.
  4. Enable the features you need in the Settings UI – everything is togglable per‑app.

Who might benefit

  • Users who want a keyboard with no data‑leaking permissions but still need AI‑assisted proofreading, translation, or voice input.
  • Developers or privacy‑conscious organisations that run their own LLM servers and want to connect a mobile keyboard to them.
  • Anyone who prefers offline functionality (e.g., on‑device Whisper or GGUF models) for work in low‑connectivity environments.

Community & extensibility LeanType ships a small plugin ecosystem (AI, voice, OCR, translation, handwriting, sound packs, themes). Each plugin is a separate APK that the keyboard loads via Android’s IPC mechanism, making it easy to add or replace components without rebuilding the whole app.

License & support

  • GPL‑v3 licensed, with sponsorship options via GitHub Sponsors and Open Collective.
  • Regular releases, a built‑in self‑updater (standard flavor), and a public issue tracker for bugs and feature requests.

Bottom line – LeanType is a genuine, feature‑rich Android keyboard that blends privacy‑first design with optional AI‑powered assistance, offering both cloud and fully offline operation.

Related

  • Project
    DevEmperor/DictateKeyboard

    An AI-powered Android keyboard that enables high-accuracy dictation and transcription across any app using customizable speech engines and on-device models.

  • Project
    ahpxex/read-aware

    ReadAware is an open‑source, cross‑platform ebook reader that embeds a persistent AI agent. The agent can retrieve passages, keep a memory across books, and answer in‑context questions, while plugins extend both the UI and the agent’s toolset. Sync is end‑to‑end encrypted via a relay or self‑hosted WebDAV, and the app works fully offline. It’s built with React, Tauri, and a Bun/Turborepo monorepo, licensed under AGPL‑3.0 (plugin API MIT).

  • Project
    theJayTea/WritingTools

    Writing Tools is a free, open‑source desktop app for Windows, Linux and macOS that lets you invoke an LLM anywhere on your computer with a hot‑key. It can proofread, rewrite, change tone, translate, summarise web pages or YouTube transcripts, and even chat with the model. You can use cloud APIs (Gemini, OpenAI, Anthropic, etc.) or run local models via Ollama, llama.cpp or Apple‑silicon MLX. The app is privacy‑focused (no telemetry, local storage of API keys) and licensed under GPL‑v3.

  • Project
    ideaplexa/voicetypr

    Voicetypr is an open‑source, offline‑first dictation app for macOS and Windows. It captures speech via a global hot‑key, transcribes locally with Whisper (and Parakeet on Apple Silicon) or optionally via cloud STT services, can clean up the text with LLMs, and inserts the result at the current cursor. A companion CLI lets scripts or AI agents use the same transcription pipeline, returning plain text or JSON. The app is built with Tauri (Rust backend) and React, offers LAN‑based private transcription servers, and is released under AGPL‑3.0.

  • Project
    niedev/RTranslator

    RTranslator is a free, offline, real-time translation app for Android. It runs two AI models (Meta's NLLB for translation, OpenAI's Whisper for speech recognition) entirely on-device, so no internet is needed and privacy is preserved. It has three modes: Conversation (two phones connected via Bluetooth translate a live dialogue in both directions), WalkieTalkie (one phone, two people taking turns), and classic text translation. It requires at least 6GB of RAM, supports 31 languages at full quality plus more at reduced quality, and is still in beta with a known Bluetooth-drop bug. Version 3.0 is in development with better translation models and full open-source status.