Skip to content

FEATURES.md

Latest commit

 

History

History
384 lines (290 loc) · 23.7 KB

FEATURES.md

File metadata and controls

384 lines (290 loc) · 23.7 KB

LeanType Features Guide

LeanType integrates with AI providers to offer advanced proofreading and translation capabilities directly within the keyboard. This guide explains how to set up the supported providers.

Index

Section Description
🆕 Summary of New Features Quick overview of what's new.
🤖 Supported AI Providers Start here to choose your AI backend.
⚡ Groq Ultra-fast cloud inference.
🌟 Google Gemini Reliable general-purpose AI.
⚙️ HF/OpenAI-compatible Use Mistral, DeepSeek, OpenRouter, etc.
🧠 Custom AI Keys Configure custom prompts and personas.
🛡️ Offline Proofreading Privacy-first, on-device AI.
👐 Two-thumb Typing (experimental) Mix taps and swipes naturally; manual spacing; Nintype-style.
📝 Text Expander Custom text shortcut expansion.
🖱️ Touchpad Mode Full-screen touchpad gestures and controls.

Summary of New Features

Feature Function Settings Location
Multi-Provider AI Uses Gemini, Groq, or OpenAI to proofread/rewrite text. Fetch latest models dynamically. AI Integration > Set AI Provider
Offline Proofreading Private, on-device AI for grammar (requires downloads). AI Integration > Offline Proofreading
Custom AI Keys 10 toolbar keys with custom prompts, tags (themed capsules), and toggle settings (supports hashtags). AI Integration > Custom Keys
AI Translation Translates selected text via your configured AI provider (includes separate model selector). Toolbar > Translate Icon
Floating Keyboard Detach the keyboard into a draggable window with a persistent mode option. Toolbar > Floating Keyboard
Touchpad Mode Swipe up on Spacebar to control cursor, including full-screen laptop-style touchpad. Gesture typing > Vertical spacebar swipe
Split Suggestions Separates suggestions from toolbar for quicker access. Appearance > Split toolbar & suggestions
Build Variants Choose Standard, Offline, or Offline Lite versions. GitHub Releases
Clear Incognito Shows a clear "Hat & Glasses" icon when Incognito is active. Automatic (when Incognito)
Clipboard Search Search history, undo swipe-delete, and optionally fold pinned items by default. Clipboard Toolbar > Search Icon
Dictionary Import Import personal words from Google Gboard/other keyboards. Text correction > Dictionary > Import
Force Auto-Caps Toggle to ensure automatic capitalization works reliably. Text correction > Auto-capitalization
Emoji Search Search for emojis by name. Emoji Key > Search Icon
Screenshot Suggestion Suggests recently taken screenshots for quick sharing. Text correction > Suggest recent screenshots
Screenshot on Clipboard Automatically saves taken screenshots to your clipboard history. Automatic (when enabled)
Clipboard Undo Undo swipe-to-delete on clipboard items with a timed undo bar. Automatic (on swipe delete)
Two-thumb Typing Mix taps and swipes naturally, multi-tap then swipe, manual spacing, recognition tweaks. All experimental and opt-in. Two-thumb typing (experimental)
Text Expander Expand custom shortcuts using dynamic template variables (date, time, clipboard, custom placeholders). Text correction > Text Expander

Emoji Search

  • Functionality: Search for emojis by keyword/name directly from the emoji palette.
  • Requirement: You must load an Emoji Main Dictionary.
  • How to Setup:
    1. Go to Settings > Text correction > Dictionary.
    2. Tap on your language (e.g., English).
    3. Ensure the Emoji English (or relevant emoji dict) is available/downloaded.
    4. Note: Basic emoji search requires this dictionary to map words to emoji characters.

Screenshot Suggestion

  • Functionality: When you take a screenshot, LeanType detects it and shows a suggestion in the suggestion strip for quick sharing. Tapping the suggestion inserts the screenshot as an image into the current input field (if the app supports image insertion).
  • Permissions: Requires storage/media access (READ_MEDIA_IMAGES on Android 13+, or READ_EXTERNAL_STORAGE on older versions). The keyboard will prompt for permission when this feature is first enabled.
  • How to Enable (Disabled by default):
    1. Go to Settings > Text correction.
    2. Enable Suggest recent screenshots.
  • Behavior:
    • Screenshots taken within the last 4 minutes are detected.
    • A thumbnail preview is shown in the suggestion strip.
    • Tapping the suggestion inserts the image; tapping the close button dismisses it.

Screenshot on Clipboard

  • Functionality: Taken screenshots are automatically saved to your clipboard history for later use.
  • Requirement: Clipboard history must be enabled.
  • How to Enable: Ensure Suggest recent screenshots is enabled in Text correction.

Clipboard Item Delete Undo

  • Functionality: When you swipe to delete an item from your clipboard history, a brief undo bar appears, allowing you to restore the deleted item.
  • Behavior: Provides a safety net against accidental deletions of important clipboard content.

Supported AI Providers

Provider Privacy Setup Free Tier Best For
Groq 🟡 Average 🟢 Easy High Speed
Google Gemini 🔴 Low 🟢 Easy Generous General Purpose
HF/OpenAI-compatible ⚙️ Varies 🟡 Medium Varies Fully Customizable
Offline (ONNX) 🟢 Best 🟡 Medium ∞ Unlimited Privacy

Tip

The HF/OpenAI-compatible option is fully customizable—you can change the API endpoint, token, and model to use any OpenAI-compatible service (OpenRouter, Mistral, DeepSeek, HuggingFace, etc.).

Privacy Links:

Important

Sensitive Data: Do not process sensitive information (passwords, credit card numbers, private addresses) using the AI Proofreading or Translation features.

Offline Version Guarantee: The Offline Version physically excludes all network code at build time. It is impossible for it to connect to the internet, making it safe for all data.

1. Groq

Groq is a cloud API provider that uses custom LPUs (Language Processing Units) to deliver extremely fast inference speeds, making it feel almost instant compared to other cloud providers.

Setup

  1. Get API Key: Visit Groq Console and create a key (starts with gsk_).
  2. Configure in LeanType:
    • Provider: Select Groq.
    • API Token: Paste your Groq API Key.

Available Models

Model ID Context Description
llama-3.3-70b-versatile 128k Best Overall. High intelligence.
llama-3.1-8b-instant 128k Fastest. Quick grammar fixes.
meta-llama/llama-4-scout-17b-16e-instruct 128k New scout model. Good balance.
meta-llama/llama-4-maverick-17b-128e-instruct 128k New maverick model. Better reasoning.
qwen/qwen3-32b 128k Good speed and logic.
openai/gpt-oss-120b 128k Large OSS GPT model.
openai/gpt-oss-20b 128k Fast OSS GPT model.
groq/compound 128k Deep-thinking internal model.
groq/compound-mini 128k Optimized internal model.
moonshotai/kimi-k2-instruct 128k Lightweight reasoning model.
moonshotai/kimi-k2-instruct-0905 128k Specialized kimi instruction model.
canopylabs/orpheus-v1-english 128k CanopyLabs English priority model.
canopylabs/orpheus-arabic-saudi 128k CanopyLabs Arabic dialect model.
allam-2-7b 128k Efficient 7B general use model.

2. Google Gemini

Get an API Key

  1. Go to Google AI Studio.
  2. Click Create API Key.
  3. Copy the key.

Configuration

  1. Go to Settings > AI Integration > Gemini API Key.
  2. Paste your API key.
  3. Select a model.

Available Models

Model ID Description
gemini-2.5-flash Fastest. Great for quick tasks.
gemini-2.5-pro High intelligence. Best overall quality.
gemini-flash-latest Fast and capable (latest flash).
gemini-flash-lite-latest Lightweight flash variant.
gemini-pro-latest Latest pro model.
gemini-3.1-pro-preview Next-gen pro preview.
gemini-3.1-pro-preview-customtools Pro preview with custom tools support.
gemini-3-pro-preview Gemini 3 pro preview.
gemini-3-flash-preview Gemini 3 flash preview.
deep-research-pro-preview-12-2025 Deep research specialized model.
gemma-3-27b-it Large Gemma model. High quality.
gemma-3-12b-it Mid-size Gemma model. Good balance.
gemma-3-4b-it Compact Gemma model. Fast.
gemma-3-1b-it Smallest Gemma model. Ultra-light.
gemma-3n-e4b-it Efficient Gemma variant (4B).
gemma-3n-e2b-it Efficient Gemma variant (2B).

3. HF/OpenAI-compatible (Generic Provider)

This provider supports any service using the standard OpenAI Chat Completion API format.

A. HuggingFace Inference API

Setup

  1. Get Token: Go to HuggingFace Settings and create a 'Read' token.
    • Note: HuggingFace may require you to add a valid payment method to your account to "unlock" the Inference API, even for the free tier (to prevent abuse).
  2. Configure in LeanType:
    • Provider: Select HF/OpenAI-compatible.
    • API Token: Paste your HF Access Token.
    • API Endpoint: https://api-inference.huggingface.co/models/<USER>/<MODEL>/v1/chat/completions
    • Model Name: <USER>/<MODEL> (e.g., meta-llama/Meta-Llama-3-8B-Instruct).

B. OpenRouter / Other Providers

  1. API Endpoint: Enter the provider's completion URL.
    • OpenRouter: https://openrouter.ai/api/v1/chat/completions
    • DeepSeek: https://api.deepseek.com/chat/completions
    • Mistral: https://api.mistral.ai/v1/chat/completions
    • OpenAI: https://api.openai.com/v1/chat/completions (Default)
  2. API Token: Enter your API Key from that provider.
  3. Model Name: Enter the exact model ID from the provider's documentation (e.g., deepseek-chat, mistral-large-latest, gpt-4o-mini).

4. Custom AI Keys & Keywords

You can assign custom prompts to 10 specific keys in the toolbar. These keys can "act" differently based on the keywords (hashtags) you include in your prompt.

How to Use

  1. Go to Settings > AI Integration > Custom Keys.
  2. Tap a key (1-10) to configure it.
  3. Enter your instructions. You can use the following hashtags to control the AI's behavior and output format.

Custom Text Labels / Tags

You can assign custom text labels (up to 12 characters) to toolbar keys instead of displaying generic AI icons:

  1. Fill in the Tag / Label field when configuring a custom AI key (e.g. French, Rewrite, Reply).
  2. Toggle Show tags on keyboard under Settings > AI Integration > Custom Keys.
  3. Custom labels will display as sleek, themed capsules directly on the keyboard toolbar.

AI Persona Keywords

Add these to your prompt to enforce a specific role.

Keyword Persona / Behavior System Instruction Added
#editor Text Editor "Output ONLY the edited text. Do not add any conversational filler."
#outputonly Strict Output "Output ONLY the result. Do not add introductions or explanations."
#proofread Proofreader "Fix grammar and spelling errors. Output ONLY the fixed text."
#paraphrase Rewriter "Rewrite using different words while keeping the meaning."
#summarize Summarizer "Provide a concise summary."
#expand Writer "Expand on the text with more details."
#toneshift Tone Adjuster "Adjust the tone as requested."

Two-thumb Typing (experimental)

LeanType ships a set of opt-in tweaks for typing with two thumbs at once — mixing taps and swipes naturally, getting fewer dropped letters when one thumb taps while the other glides, and Nintype-style "manual spacing" where the word doesn't auto-commit until you tap space. All options live under Settings → Two-thumb typing (experimental) and default to off (or 0 for the sliders), so you only get the behaviour you opt into.

⚠️ These features change how gestures and taps interact. Try them one at a time so you can tell what's helping vs. hurting your typing.

Spacing & word commit

Pref Behaviour
Manual spacing (gesture_manual_spacing) When ON, lifting all fingers after a gesture does not auto-commit or insert an autospace. The gesture's result becomes/extends a composing word. Multiple gestures and taps chain into one logical word until you explicitly tap space, punctuation, or Enter. Best paired with the next option.
Backspace removes last fragment (gesture_fragment_backspace) Only visible when Manual spacing is on. Backspace pops the entire previous fragment (one swipe's output, or one tap's letter) in a single keystroke instead of one character at a time.
Autospace grace period (gesture_autospace_grace_ms, 0–500 ms) Only visible when Manual spacing is off (mutually exclusive). A middle ground: keep the autospace, but delay it by N ms after the last finger lifts. If a new finger goes down within that window, the deferred commit is dropped and the gesture continues into the same word — no matter how long that next gesture itself takes. Set to 0 to disable. While the timer is pending, the floating preview text shows a trailing ellipsis (…) so you can see the system is waiting.

Side effect: when Manual spacing is on, the Autospace before/after gesture typing toggles under Settings → Text correction → Space are hidden — they become runtime no-ops.

Combining taps and swipes

Pref Behaviour
Tap-then-swipe window (gesture_tap_promotion_ms, 0–200 ms) When > 0, lets you tap one or more letters then immediately swipe the rest, getting one merged word. Example: tap p, tap a, swipe ul → paul. Set to 0 to disable. The window is measured against the last letter input — chains grow naturally as long as each new tap arrives within the window of the previous one.

Visual feedback

Pref Behaviour
Flash space key on auto-space (autospace_visual_hint, on by default) When ON, the space bar briefly highlights every time an automatic space is silently inserted. Most useful when grace-period auto-spacing is on, since the actual insertion is decoupled from any visible keystroke.

Layout extras

Pref Behaviour
Apostrophe key for gestures (gesture_apostrophe_key) When ON, surfaces a swipeable apostrophe key so contractions like it's or don't can be glided in one stroke. Toggle is visible; actual layout integration depends on whether the active layout exposes the apostrophe — see layouts.md if you maintain a custom layout.

Recognition & debug

Pref Behaviour
Two-thumb point hinting (experimental) (gesture_dual_thumb_hinting) When ON, the keyboard post-processes the gesture's raw points before feeding them to the recognizer: tap-bursts that overlap an active stroke get reinforced with synthetic on-stroke waypoints at the tap's centre so the library doesn't under-weight them. Includes a proximity guard so a stray opposite-hand tap doesn't get amplified into a recognition-deforming detour. Most useful for words like firetruck (fretrc swipe + iuk taps).
Left/right hand split (gesture_dual_thumb_midline_pct, 30–70 %) Only visible when Point hinting is on. Where the keyboard splits between your left and right hand. Currently used only by the (forthcoming) stray-tap dampener; the proximity guard above is independent of this slider.
Draw gesture points (debug) (gesture_debug_draw_points) When ON, overlays the raw gesture samples (small red dots) and any synthetic points injected by Point hinting (larger blue dots) on the keyboard. Useful when iterating on hinting or filing bug reports — turn off for daily typing.

Why these exist

These features all started from HeliBoard issue #291 ("Improving simultaneous/two-finger swiping") and Nintype's old "autospace off" + "tap during swipe" behaviours. They're shipped as experimental opt-ins so you can A/B them on your own typing style without risk to the existing single-thumb gesture experience.

| #generate | Content Generator | "You are a creative content generator. Output ONLY content." |

Input Handling Keywords

Control how the result is inserted.

Keyword Behavior Use Case
(Default) Replace: The AI output replaces the selected text or the entire text field content. Proofreading, rewriting, summarizing.
#append Append: The AI output is added to the end of the text field (or selection) instead of replacing it. Generating replies, continuing a story, adding a sign-off.
#showthought Show Thinking: Preserves the "thinking" process (e.g., <think>...</think>) from reasoning models like Qwen. Debugging reasoning, seeing the AI's thought process.

Examples

1. Standard Proofreading (Replace)

Prompt: Fix grammar #proofread Result: Replaces your text with the corrected version.

2. Generate a Reply (Append)

Prompt: Generate a polite decline to this invitation #generate #append Result: Keeps the original invitation text and adds your polite decline at the end.

3. Strict Rewriting

Prompt: Rewrite this to be more professional #editor Result: Replaces text with professional version, guaranteeing no "Here is the text:" prefixes.

4. Tone Shift

Prompt: Make this sound more enthusiastic #toneshift Result: Rewrites the text with high energy and exclamation points, keeping the core message.

Tip

Hashtags are Optional: You can write purely custom prompts (e.g., "Translate to French"). However, without a hashtag like #outputonly or #editor, the AI might act like a chatbot (e.g., responding with "Sure! Here is the translation: ..."). Using these keywords automatically injects strict system instructions to ensure you get only the result you want.

Privacy

  • Data: Text is sent directly from your device to the chosen API provider. No intermediate servers are used.

5. Offline Proofreading (Privacy Focused)

Note: This feature is only available in the "Offline" build flavor of LeanType.

Offline proofreading runs entirely on your device using the ONNX Runtime engine. No data leaves your device.

Note

Status: Beta / Experimental This feature is in a test phase. The engine is designed to be compatible with various T5-based ONNX models (Basic, Quantized, KV-Cache). We encourage you to experiment with different models to find the best balance of speed and accuracy for your device.

Setup Instructions

  1. Download Model Files: Download the Encoder, Decoder, and Tokenizer for your chosen model from the table below.
  2. Configure App:
    • Go to Settings > Advanced.
    • Encoder Model: Select the downloaded .onnx encoder file.
    • Decoder Model: Select the downloaded .onnx decoder file.
    • Tokenizer: Select the tokenizer.json file.
    • System Instruction: Enter the text specified in the "System Instruction" column for your model (leave empty if specified).

Recommended Models

Model & Purpose Performance / Size System Instruction Download Links (Direct)
Visheratin T5 Tiny
(Grammar Correction Only)		 ⚡ Fastest
~35 MB
Low RAM usage		 Empty
(Leave blank)		 • Encoder
• Decoder
• Tokenizer
Flan-T5 Small
(Translation & General)		 🐢 Slower
~300 MB
Higher accuracy		 Required
fix grammar:
or
translate English to Spanish: 		• Encoder
• Decoder
• Tokenizer

Note: For Flan-T5, the quantized models linked above are standard recommendations. Users have also reported success with bnb4 quantized variants if available.

6. Text Expander

Text Expander allows you to define custom shortcuts (abbreviations) that automatically expand into longer, structured text templates as you type.

Key Features

  • Custom Shortcuts: Create abbreviations (e.g., adr) that expand immediately into complex blocks (e.g., your full postal address).
  • Dynamic Template Variables: Enrich your expansions using dynamic variables:
    • %date% - Inserts the current local date.
    • %time% - Inserts the current local time.
    • %clipboard% - Appends the most recently copied text from your clipboard.
    • %cursor% - Positions the typing cursor here after expansion.
    • %greeting% - Inserts "Good morning", "Good afternoon", or "Good evening" depending on the hour.
    • %tomorrow% - Inserts tomorrow's date (YYYY-MM-DD).
    • %bullets% - Inserts a bullet list template (supports count suffix e.g. %bullets_5%).
    • %list% - Inserts a numbered list template (supports count suffix e.g. %list_5%).
    • Custom Placeholders: Create dynamic input fields (e.g., %name%) that prompt you to type a value during the expansion flow.

Configuration

  1. Navigate to Settings > Text correction > Text Expander.
  2. Tap the + (Add) button to create a new expansion rule.
  3. Specify the Shortcut trigger and the Expansion template.
  4. Include dynamic template variables in the template block.

7. Touchpad Mode

Touchpad Mode replaces the keyboard with a laptop-style touchpad overlay to control the cursor and edit text using fluid gestures.

How to Enable

  • Swipe gesture: Swipe up on the Spacebar to temporarily toggle Touchpad Mode.
  • Toolbar shortcut: Tap the Touchpad icon in the toolbar for a persistent touchpad overlay.

Touchpad Gestures

  • Single-finger drag: Moves the cursor in 2D space (simulating arrow keys left/right/up/down) to navigate text.
  • Two-finger drag: Performs fast vertical scrolling (simulating arrow keys up/down).
  • Two-finger tap: Simulates a mouse click/Enter.
  • Long press (hold finger): Activates text selection mode. Dragging while holding will select text. Releasing the finger exits selection mode.
  • Double tap by single finger: Deletes the selected text or words (if a text selection exists).