Changelog — AskWatch

Version 2.2.43 Build 731 Latest October 2026

✨ New Features

Maps and location — tell your AI where you want to go and it opens Apple Maps for you, and it can use where you are to answer; works in chat and voice on iPhone, Apple Watch, Mac and in CarPlay, and your location is shared only when you allow it
Azure AI Speech — a new voice for reading answers aloud on iPhone, CarPlay, Apple Watch, Apple TV and Mac, with your own Azure key, region and voice
Gemini 4 Argon, Google's newest model, with your own Google key or with AskWatch Ultra
Notes work with more AI models — search and save your notes with models from xAI, Mistral, DeepSeek, Groq, OpenRouter, Perplexity and more, using your own key
iPhone Duo — Chat, Home, Notes, Settings, Compare and Live Conversation can use two columns on the open screen
Custom AI providers get a Capabilities section with a test button that shows what your server supports, on iPhone and Mac

Version 2.2.42 Build 724 October 2026

✨ New Features

Claude Max includes monthly API credits ($100 on Max 5x, $200 on Max 20x) — link a Claude Console organization and use Claude in AskWatch with your own key at no extra API cost
Piper Cloud voices now read answers in Japanese and Thai, included with PRO, AskWatch+ and Ultra with no key of your own

Version 2.2.41 Build 713 October 2026

✨ New Features

Claude Sonnet 5.5 and GPT-6.1 Sol added
When an Ultra daily allowance for a model runs out, AskWatch can continue the answer on that provider's fast model, with a short note, instead of stopping

🔧 Improvements

Apple Watch Voice Chat now shows an error message and vibrates when transcription runs out of time, instead of silently dropping your question

Version 2.2.40 Build 712 September 2026

🔧 Improvements

Apple Watch Medium and Compact layouts use larger tap areas of at least 44 pt, and on the smallest watches Templates is an icon next to the main button
Perplexity with your own key now uses Perplexity's new Agent API, with web sources as before

🐛 Bug Fixes

The Apple Watch Compact layout shows a "Listening" label while Voice Chat listens, instead of an empty screen
Large Voice Chat fits 40 and 41 mm Apple Watches, with the status and the controls kept on screen

Version 2.2.39 Build 710 September 2026

🔧 Improvements

Apple Watch Compact layout: the Skip button in Voice Chat and the Stop button in Speak take a tap in a 44 × 44 pt area

🐛 Bug Fixes

OpenAI Live answers again after you press Skip; before, it stayed silent until you closed the session
On small Apple Watches the Large layout no longer overlaps: the Templates button covered the provider name, and on the Home Assistant page the main button ran under the page dots
The web-search icon on the Apple Watch home screen now shows whether your question will really be searched
Home Assistant alerts are no longer sent when the entity's state did not change, such as "Idle to Idle"

Version 2.2.38 Build 707 September 2026

✨ New Features

OpenAI Live, the first full-duplex voice model in AskWatch — it listens while it speaks, so you can interrupt it, and it can search the web, control your home and save notes; 13 voices, three GPT-5.6 backend models and your own opening greeting, with your OpenAI key, on Apple Watch, iPhone, Mac and in CarPlay
Claude Opus 5.5, GPT-6 Sol and GPT-6 Luna added; GPT-6 Luna is also included in AskWatch Ultra
Conversation Layout on Apple Watch — Automatic, Large, Medium or Compact, so smaller watches get more room for the text
Apple Watch complications work again on watchOS 10 and 11 (Series 4, Series 5 and SE 1st generation)

Version 2.2.37 Build 689 September 2026

✨ New Features

Gemini 3.8 Live for Live Conversation with Google — used automatically on a subscription, or pick it with your own Google key; it keeps Home Assistant control, notes and web search
Web search in Live Conversation — Gemini Live and Grok Voice can search the web while you talk, with one switch per provider and device

Version 2.2.36 Build 681 September 2026

✨ New Features

Notes — a private notebook your AI can search, save to and edit by voice, on iPhone, Apple Watch, Mac and in CarPlay, with the models that support it; switch it on in Real-Time Context
Export all your notes to one plain-text file from the notes list on iPhone and Mac, and hide a single note from the assistant
Reset Everything now asks how far the reset should reach

Version 2.2.35 Build 673 September 2026

✨ New Features

Z.ai joins as the 14th AI provider — GLM 4.5 through 5.3 with your own key, and GLM 4.7 in AskWatch+ and GLM 5.2 in Ultra with no key at all
Grok 4.6 and Grok 4.3 added; three retired Grok models now resolve to 4.3 automatically
Perplexity's Sonar models available through OpenRouter, so Apple Watch can pick them the same way iPhone can
Live Conversation can open with a greeting you write yourself — turn on "Assistant speaks first" for your provider; Apple Watch has its own setting and CarPlay follows your iPhone
Sources under web answers — the numbered pages behind an answer, with Perplexity, OpenRouter, Anthropic and OpenAI

Version 2.2.34 Build 658 August 2026

✨ New Features

AskWatch for Mac arrives on the Mac App Store — menu bar app, floating quick-ask panel, Safari extension, command-line tool and a native Home Assistant dashboard
Mac gets its own Help window (⌘?) with a searchable sidebar and 16 Mac-specific topics
Mac Settings gain a Safari tab to configure the web extension, its language and its context-menu items
Shortcuts return real answers on Mac and iPhone — Ask AI, Ask Question, Translate and Summarize all work in the background
Mac dictation can stop by itself after a pause you choose, send automatically once transcribed, and append to a draft instead of replacing it
Live Conversation on Mac no longer starts on its own — a Start button, in-window provider switching and a separate Stop control
OpenAI joins Google as a second managed Live Conversation engine for AskWatch+ and Ultra
AskWatch+ daily Live Conversation time raised from 10 to 15 minutes

Version 2.2.33 Build 639 August 2026

✨ New Features

Gemini 3.7 Flash added and set as the new Gemini default
Apple Watch gets its own Model picker under AI Provider, kept per watch instead of being overwritten from iPhone
Apple Watch separates system volume on the Digital Crown from the spoken-answer volume
Voice preview in Settings — new on Mac, and on Apple Watch it now covers all six speech engines instead of Apple's only
Apple TTS volume slider added on iPhone, Mac, Apple TV and CarPlay
The Mac command-line tool gains --lane, --providers and --status
Mac offers a city picker when it cannot determine your location

Version 2.2.32 Build 622 August 2026

✨ New Features

Attachments work everywhere: PDF documents as well as images, an Attach File button on iPhone, and attachments finally reach managed subscriptions and custom providers
Live captions while you speak, through OpenAI's live transcription
Piper Cloud adds Russian, Korean, Hindi, Indonesian and Hebrew voices — free, no API key
Apple TV gets a thinking-level setting of its own
DeepSeek V4 Flash added to the managed line-up
New 'Open Chat on launch' option on the iPhone Home tab

Version 2.2.31 Build 611 July 2026

✨ New Features

Currency picker for exchange rates in Settings, followed automatically by Apple TV
Claude Opus 5, Grok 4.5, Gemini 3.6 Flash and Gemini 3.5 Flash-Lite added
The Mac command-line tool becomes installable from Settings and can use your managed subscription, not only your own key
Mac quick-ask panel gains a provider switcher, a position setting and an 'Open in app' action
Mac Settings gain a Support section and a 'Launch minimized' option

Version 2.2.30 Build 608 July 2026

✨ New Features

Lifetime PRO can be bought on Mac — previously it could not be purchased there at all
Mac Home Assistant reaches full parity: around 23 device types with the same inspector iPhone has
Custom providers can be set as the active provider on Mac
Mac Home Assistant dashboard gains category filters and a native What's New window

Version 2.2.29 Build 605 July 2026

✨ New Features

OpenAI's GPT-5.6 models added — Sol, Terra and Luna
Home Assistant control works inside Live Conversation, hands-free, on iPhone, Apple Watch, CarPlay and Mac
Apple Watch gets its own Live Conversation settings, synced both ways with iPhone
Managed Live Conversation gains a region picker
Prompt Templates reworked — usable in Voice Chat and editable on iPhone
All 26 Grok voices available for spoken answers
New 'Continue Previous Conversation' choice for Voice Chat, off by default

Version 2.2.28 Build 595 July 2026

✨ New Features

Claude Sonnet 5 added, included in AskWatch Ultra with no API key
Live Conversation now uses your own system prompt and real-time context — date, weather and finance — instead of one generic prompt
The System Prompt setting moved out of the model screen into a section of its own

Version 2.2.26 Build 588 June 2026

✨ New Features

Live Conversation shows which microphone is active and warns when it hears nothing
OpenAI users can choose their transcription model for Live Conversation
Gemini Live voice choices grow from 8 to 30

Version 2.2.25 Build 584 June 2026

✨ New Features

Switching AI provider on Apple Watch is now a tap instead of a swipe
New FAQ explaining the difference between Dictate, Speak, Voice Chat and Live Conversation

Version 2.2.24 Build 579 June 2026

✨ New Features

Bluetooth headphones with the Apple Watch microphone leave beta
New 'Interrupt response' option so you can talk over an answer
New Context 'Off' choice — send only the current message, with no history — available on the free tier

Version 2.2.23 Build 568 June 2026

✨ New Features

Managed Live Conversation — AskWatch+ and Ultra can hold a live voice conversation through Gemini with no API key of their own
Apple Watch Live Conversation gains history, a Reconnect button and a clear message when a key is missing
Interrupting an answer works with headphones and AirPods

Version 2.2.22 Build 555 June 2026

✨ New Features

Live Conversation launches — a real back-and-forth voice conversation with OpenAI, xAI or Google on iPhone, Apple Watch and CarPlay
Live Conversation comes to Mac with its own window and a ⌘5 shortcut
A native Home Assistant dashboard on Mac: three layouts, live charts, a camera wall and an entity inspector
The Mac app redesigned for macOS 26, with a native onboarding wizard and one-tap demo setup
Five new Grok voices for spoken answers
Apple Watch Ultra's Action Button can open straight into a chosen voice mode

Version 2.2.21 Build 530 May 2026

✨ New Features

Voice Chat gets its own Action Button shortcut
The Shortcuts app on iPhone can build Speak and Dictate shortcuts for the first time
Action Button setup becomes a guided three-step flow

Version 2.2.19 Build 519 May 2026

✨ New Features

Gemini 3.5 Flash and Gemini 3.1 Flash-Lite added

Version 2.2.15 Build 506 May 2026

✨ New Features

Voice quality badge — Apple voices now show a Premium or Enhanced label so you can pick the best-sounding voice; clearer guidance about Siri voices
Apple TV — read aloud toggle so you can choose between voice and text-only answers
Compare Providers History — review past comparisons across all your AI providers (Plus subscribers)
OpenRouter Web Search — ask any OpenRouter model to search the web, with a 1–10 slider to control depth
GPT-5.5 and GPT-5.5 Pro — OpenAI's latest models added to the curated list
Gemma 4 (26B and 31B) — Google's open models added
MiniMax M2.7 — replaces an older DeepSeek model on Together AI

🔧 Improvements

Subscription badge — see your tier (Free, Plus, Ultra, or PRO) and renewal date everywhere it matters
Cleaner read-aloud — markdown tables and unpronounceable text are stripped before being spoken
Spoken quota messages — CarPlay and Apple Watch now say quota or subscription warnings out loud instead of failing silently
Refined Watch typing flow when sending prompts and using Home Assistant commands
More reliable subscription state across iPhone, Watch, CarPlay, and Apple TV

🐛 Bug Fixes

Fixed Apple Watch occasionally leaking Home Assistant prompts when HA was disabled
Fixed CarPlay 'No favorites configured' appearing untranslated in some languages
Restore Purchase analytics now distinguish cancelled vs failed restores
Various stability and CarPlay audio improvements

Version 2.2.6 Build 408 April 2026

✨ New Features

Custom Watch screens with custom providers — pin your own AI server to a Watch shortcut
Custom ElevenLabs voices — use your cloned or designed Voice ID for ultra-personal text-to-speech
PRO offer code redemption — redeem promo codes directly inside the app
Reset Everything — PIN-protected hard reset on iPhone and Apple Watch for shared devices
Quick Start with Demo — instant configuration for first-time users and reviewers

🔧 Improvements

AI providers grouped as Recommended, Verified, and Other for easier model picking
Home Assistant context toggle now works for custom providers too
Configurable token parameter for custom providers (max_tokens vs max_completion_tokens)
Helpful info text for TTS and STT when another custom provider is active

🐛 Bug Fixes

Fixed several edge cases in Apple Watch subscription verification
Fixed Apple Weather attribution on Apple TV settings
More resilient Restore Purchase flow across the whole app

Version 2.0.18 March 2026

✨ New Features

Apple TV app — ask AI on the big screen using your iPhone as a wireless microphone
CarPlay — hands-free voice AI and smart home control while driving
Web Search — AI can now search the web for up-to-date answers
Custom Provider — connect your own AI server (Ollama, vLLM, or any OpenAI-compatible endpoint)
Live Activities — monitor your smart home sensors on iPhone Lock Screen and Dynamic Island
Weather & Finance — AI answers now include current weather, exchange rates, stocks, and crypto prices
Custom voice engines — choose your preferred text-to-speech and speech recognition provider

🔧 Improvements

Now available on 5 Apple platforms — iPhone, Apple Watch, iPad, CarPlay, and Apple TV
Smarter smart home — AI understands your question language and shows only relevant devices
Push notifications for smart home — get alerts even when the app is closed
Multiple Home Assistant servers — connect more than one smart home with automatic switching
Better voice recognition — automatic retry if transcription fails

🐛 Bug Fixes

Fixed a crash when deleting conversations quickly
Fixed smart home incorrectly showing "unreachable" during high traffic
Fixed photo attachment permission issue
Various stability and performance improvements

Version 2.0.5 Build 128 February 2026

✨ New Features

Smart Home Dashboard — control Home Assistant devices directly from iPhone and Apple Watch. Lights, climate, covers, cameras, vacuums, scenes, automations, and more
AI-Enhanced Smart Home — control your home with natural language ("turn on living room lights", "set bedroom to 22°C")
4 new AI providers — Cerebras, Cohere, Together AI, and SiliconFlow (13 total)
Compare Providers — compare AI responses side by side from multiple providers at once
Custom Watch Home Screens — create personalized Watch screens with different providers and modes
Cloud Text-to-Speech — high-quality OpenAI TTS with 9 voice options (alloy, coral, nova, and more)
In-app language selector — switch interface language without changing device settings
Remote Control — full TV/media remote in entity detail (D-pad, volume, channels, sources)

🔧 Improvements

Direct Watch transcription — Voice Chat and Speak AI work independently without iPhone nearby
32 language localizations — expanded from 10 to 32 languages (1450+ translated strings)
20+ smart home entity types — lights, climate, covers, cameras, vacuums, water heaters, valves, humidifiers, timers, counters, lawn mowers, and more
Smart Home dual URL — local network + Nabu Casa cloud with automatic fallback
Cloudflare Tunnel support — connect to Home Assistant behind Cloudflare Access
Optimized dashboard scrolling with debounced updates and reduced rebuild frequency

🐛 Bug Fixes

Fixed Voice Chat audio buffer race condition
Fixed App Attest double encoding issue
Fixed TTS restart and audio session issues
Fixed brightness slider snap-back in light cards
Improved HTTP timeouts for slower AI models

Version 2.0 Build 109 January 2026

✨ New Features

Voice Chat Mode - hands-free continuous conversation with automatic speech detection
3 Voice Modes: Dictate AI, Speak AI (push-to-talk), Voice Chat (hands-free)
iOS Chat - full multi-turn conversations on iPhone with encrypted storage
6 Watch complications for quick access to any voice mode
3 Control Center buttons on watchOS
Interactive iOS Widgets (Small, Medium, Large, Lock Screen)
Tips section with 13 illustrated guides
Voice mode toggles in Settings (enable/disable individual modes)
Auto language detection from AI responses for TTS
Handoff - continue conversations from Watch to iPhone
Context limit setting (configurable message history)
History search on iOS and Watch

🔧 Improvements

AES-GCM encryption for chat history and app configuration
Unified Keychain access (iOS + Watch)
Dynamic VAD (Voice Activity Detection) with adaptive noise floor
Cloudflare Worker transcription with language detection
TTS audio session improvements for watchOS
iPad Chat UI with better sidebar handling
OpenAI API compatibility for o-series models
Deep links for complications and Control Center
600+ localized strings in 10 languages

🐛 Bug Fixes

Fixed Voice Chat audio buffer race condition
Fixed Voice Chat dead mic issue (Bluetooth routing)
Fixed feedback form file attachment issues
Fixed widget tap-to-open reliability
Fixed TTS not starting after recording

Version 1.1 Build 45 December 2025

✨ New Features

Action Button support (Watch Ultra)
6 built-in prompt templates (Translate, Summarize, Explain, Email, Grammar, Brainstorm)
Custom prompt templates
Siri Shortcuts integration
Export/Share conversations
Double-tap gesture support (Series 9+, Ultra 2+)
Watch face data complication with query counter
Conversation memory within sessions

Version 1.0 Build 1 December 2025

✨ Initial Release

9 LLM providers: OpenAI, Anthropic, Google, Perplexity, xAI, Groq, Mistral, DeepSeek, OpenRouter
Voice input with Text-to-Speech responses
iOS companion app for API key management
Secure Keychain storage with optional Face ID/PIN
Dynamic model loading from provider APIs
Balance/Credits API for supported providers
Watch complications
System prompt per provider
Conversation history (iOS + Watch)
10 language localizations