1.8

Update

New Features

No new features in this release.

Bug Fixes

• Safely drain model tasks before shutdown
• Android Studio ghost text font and vertical alignment
• Pages ghost text baseline and find alignment
• Visual Studio Code ghost text editor placement
• MarkEdit ghost text multiline alignment
• Safari ghost text ChatGPT paragraph alignment
• ChatWise ghost text multiline and soft wrap alignment
• Xcode ghost text find and replace alignment
• Microsoft Word ghost text font and find alignment
• Microsoft Outlook ghost text field and body alignment
• Mail ghost text search and subject alignment
• Chrome ghost text address bar suppression
• ChatGPT ghost text font sizing
• TextEdit ghost text vertical alignment

Requirements

• macOS Sonoma (14.0) or later
• Apple Silicon Mac recommended for on-device model inference

1.7

Update

New Features

• Add input method controls
• Batch correction scoring paths
• Adapt beam search work
• Promote generated frontier state
• Stage long burst decoding
• 4 token default completions
• Context aware token budgets
• Grammar correction
• Typo correction
• Add LFM2.5-8B-A1B-Base for high RAM Macs
• Enable writing history and clipboard history by default

Bug Fixes

• resolve bare detected-language tags to the user's regional spelling dictionary (#41)
• restore ModelSetupCoordinator.swift and anchor the Models/ gitignore rule (#40)
• Remove agressive correction path

Requirements

• macOS Sonoma (14.0) or later
• Apple Silicon Mac recommended for on-device model inference

1.6

Update

New Features

No new features in this release.

Bug Fixes

• Messages ghost text with attachment
• Remove tall web field ghost text suppression
• Browser ghost text tall multiline fields
• Chrome ghost text tall web fields
• Sublime Text ghost text unreliable editor placement
• Chrome ghost text vertical baseline
• Safari ghost text vertical baseline
• TextEdit ghost text vertical baseline
• Excel ghost text multiline cell suppression
• Google AI Studio ghost text multiline prompt alignment
• Browser ghost text single-line web fields
• TextEdit ghost text vertical alignment overshoot
• Google ghost text search field alignment
• Chrome ghost text lower vertical alignment
• Safari ghost text lower vertical alignment
• TextEdit ghost text lower vertical alignment
• Safari ghost text omnibox overlay
• Chrome ghost text omnibox overlay
• Terminal ghost text invisible overlay
• Chrome ghost text too high
• Safari ghost text too high
• TextEdit ghost text too high
• Pages ghost text stale probe
• Chrome ghost text inline baseline
• Safari ghost text inline baseline
• TextEdit ghost text inline baseline
• Chrome ghost text lower baseline
• Safari ghost text lower baseline
• TextEdit ghost text lower baseline
• Chrome ghost text search overflow
• Chrome ghost text multiline wrap overlap
• Chrome ghost text high baseline
• Safari ghost text high baseline
• TextEdit ghost text high baseline
• VS Code ghost text line-origin overlay
• Safari Chrome ghost text suppress multiline web drift
• Chrome ghost text lower baseline
• Safari ghost text lower baseline
• TextEdit ghost text lower baseline
• Chrome ghost text vertical alignment
• Safari ghost text vertical alignment
• TextEdit ghost text vertical alignment
• Chrome TextEdit Safari ghost text vertical alignment

Requirements

• macOS Sonoma (14.0) or later
• Apple Silicon Mac recommended for on-device model inference

1.5

Update

New Features

• Screenshot assisted ghost text alignment
• TextKit render and better font capture
• Enable re-optimized mid-line completions
• Developer rect debug overlay
• Support English variants

Bug Fixes

• Remove Discord specific path
• Improved text alignment
• Codex ghost text alignment
• Discord suggestion misalignment
• Use string injection in iA Writer and Messages
• Release script

Requirements

• macOS Sonoma (14.0) or later
• Apple Silicon Mac recommended for on-device model inference

1.4

Update

New Features

• Manual exclude app
• Settings window resize
• Latency snapshot comparison
• 1.11x speedup via zero-copy prefix checks
• 1.06x speedup via numeric mid-word suppression
• 1.18x speedup via adaptive beam width
• 1.06x speedup via guarded mid-word decode budget
• 2.39x speedup via default mid-line suppression
• 1.33x speedup via retokenized anchor history
• 1.27x speedup via overlapped debounce generation
• 1.18x speedup via tighter adaptive debounce
• 1.1x speedup via batched FIM rerank
• Improved prompt logging
• KeyTypeBench results
• Eval harness

Bug Fixes

• Slack completion insert

Requirements

• macOS Sonoma (14.0) or later
• Apple Silicon Mac recommended for on-device model inference

1.3

Update

New Features

• End to end latency profiling
• Improve non-Latin suggestions

Bug Fixes

• Notion compatibility
• Multiline text field wrapping
• Obsidian compatibility
• Slack and Discord compatibility
• Invasive accessibility interference
• Improve Chromium app focus capture
• Dynamic web app detection

Requirements

• macOS Sonoma (14.0) or later
• Apple Silicon Mac recommended for on-device model inference

1.2

Update

New Features

• Branch promotion rollback support
• Branch promotion cache

Bug Fixes

• Screencapture compatibility
• Default OCR to off
• ACPF profile build for duplicate-byte tokenizers (Gemma) (#9)

Requirements

• macOS Sonoma (14.0) or later
• Apple Silicon Mac recommended for on-device model inference

1.1

Update

New Features

• Show app icon for windows
• FIM quality improvements
• Beam search early exit

Bug Fixes

• Mid-word healing extra space
• Type-through race condition
• Native app ghost text redrawing
• ACPF profile build error logs

Requirements

• macOS Sonoma (14.0) or later
• Apple Silicon Mac recommended for on-device model inference

1.0

KeyType

On-device, system-wide tab autocomplete for macOS.

Features

• System-wide ghost-text completions accepted with Tab
• Fully on-device prediction with a local LLM (no network calls)
• Short, cursor-anchored continuations that prefer silence over a wrong guess
• App-aware insertion and overlay behavior
• Private by default: clipboard, screen/OCR, and writing-history context are opt-in

Requirements

• macOS Sonoma (14.0) or later
• Apple Silicon Mac recommended for on-device model inference