Changelog

BoltAI v1.8

This release focuses on supporting more API Providers: Anthropic, Azure OpenAI, OpenRouter...

Added support for Anthropic, Azure OpenAI & OpenRouter
Advanced model configuration:
- Custom API Endpoint & Proxy
- Custom System Instructions
- Custom GPT parameters: context limits, temperature, presence penalty, frequency penalty
Universal full-text search
Automatically switch to GPT3.5 16K if needed
Better chat managements: edit, regenerate, delete & fork messages
General bug fixes and improvements

This release focuses on stability & performance improvement

🔍 Web Search: This plugin allows BoltAI to access the Internet and search for the most accurate data in real-time.
Quickly change GPT model, temperature and other params from the chat interface (thanks @nerdymomocat)
Added support for Regenerating Response & Continue Generating (similar to ChatGPT)
Supports multiple conversations in parallel
Added cost estimation: show a total cost estimation on top of each conversation
Add the a shortcut key to activate BoltAI
Add the option to keep chat windows always Stay on Top
Added the option to toggle Markdown rendering for better rendering performance
[QoL]: allows bulk deleting AI Commands / Assistants / Prompts (thanks @nerdymomocat)
Improvement: improve scrolling performance on large chat
Improvement: improve chat management performance
Improvement: better license key management
Bug fix: fixed the issue with image generation (Stable Diffusion & OpenJourney)

New in build 21

✨ Full Chat UI: BoltAI now features a full Chat UI with conversation management and search capabilities.
Expanded support for OpenAI models like gpt-3.5-turbo-16k, and gpt-4-32k (available when the API key allows)
You can now choose to show or hide AI Commands, AI Assistant, or Prompts
Customize the trigger key for AI Assistants or Prompt Library Popup using options like "//" (double slash), "\" (backslash), or "\\" (double backslash) instead of the popular character "/"
Fixed Content Length Crash Issue: Fixed the critical issue where the app would crash if the content was too long.
Adjustable chat font size: ⌘+ (increase), ⌘- (decrease) or ⌘0 (reset). Thanks @kpxz_
Bug fix: fixed the issue where chat messages get cut off sometimes.
Bug fix: fixed the issue where the AI Command's "behavior" doesn't work as expected.
Bug fix: fixed the crash while using the trigger shortcut in Google Docs
Bug fix: attempt to fix the Language Output not functioning as expected

This release improves 2 majors area: a new Chat UI and a much better compatibility

Chat UI: Introducing the new Chat UI.
- You can now chat with AI after using an AI command, or you can start a new chat directly from the main screen.
- It supports markdown & code highlighter. You can also choose to copy a codeblock, or the whole AI answer.
- All your conversations are saved in a local database. I will add full Chat UI with list of conversations & full-text search in the next release.
Better compatibility: BoltAI now supports non-standard apps and websites such as Google Docs, ClickUp, Slack...
More customizations: You can configure output language, and GPT model for each AI Command & AI Assistant
[QoL] Template Duplicate: You can duplicate templates and make minor edits without having to copy and paste them
General bug fixes & performance improvement

Critical Bug Fix: please upgrade to this version. It fixes the issue where AI responses might be interrupted randomly.
Added the option to stream AI responses word by word. If enable, BoltAI will emulate typing on your Mac. Any interaction will immediately stop the response.

✨ AI Commands (beta): Introducing the new AI Commands. You can now select text, press Control + Space then choose an AI command.
Language output for AI Assistant: you can select the default language for an AI Assistant
Custom temperature for AI Assistant: configure the temperature param to make the answer more creative or more precise
Minor bug fixes & improvement

Added AI assistants: you can now chat with multiple AI Assistants. Just type "/a" (without quotes), search for an assistant then press Tab
Added Prompt Library: type "/p" (without quotes), search for prompt then press Tab.
Default to "chat" mode, instead of "ask" mode
Support custom shortcut: you can choose your own shortcut key other than the default Command+Shift+Return
Fixed issue where BoltAI runs without a trigger phrase

Support new apps: Apple Mail, Mimestream, iA Writer, Android Studio, TablePlus, Slack & Discord
You can now save your generated images (default: your Download folder)
Data Privacy: redact sensitive information such as email addresses, credit card numbers & phone numbers
QoL: BoltAI now includes an auto-updater