Every public release of Triplo AI, desktop and mobile, newest first.
108 public releases since 14 February 2023 — 74 desktop and 34 mobile. The newest is 7.0.5β, 15 October 2026.
Version 7.x 2026
7.0.5βDesktop
Added missing providers, including OpenAI and Anthropic
Added a rename option for history items and an icon that opens the history folder on disk
Added a hotkey for Triplo Pi (Cmd+Shift+A)
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
7.0.4βDesktop
Fixed several Triplo Pi issues
Improved folder permissions handling and fixed agent startup issues
Fixed bug on Youtube Scraper
General UX/UI improvements
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
7.0.2βDesktop
Fixed bugs in the Auto-Update process
Improved the Auto-Update process to display a modal requesting user approval before updating
UI now shows a notification when updates are available
Fixed a bug preventing the input bar from displaying properly on long prompts
Added a migration trigger (v6 → v7) under Help & About
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
7.0.1βDesktop
Awareness fixed on Windows
//model to change the LLM Model in use on any prompt
CMD+C or CTRL+C to copy the last response from AI
Enhanced security
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
7.0.0βDesktop
A Fresh new Triplo AI for you (v7)
All in MD files — SmartPrompts, Instructions, Automations and history are Markdown under ~/Documents/Triplo AI/, watched live (~0.5 s), so bulk edits, diffs and git sync finally work
Minds are now real folders on disk — an INGEST/ drop folder, two-phase processing, Markdown + YAML frontmatter output, originals preserved in files/, failures kept in failed/, bulk URL import
Semantic search that runs locally — bundled ~52 MB embedding model, no uploads, no runtime download, index updates within seconds, keyword search as automatic fallback
An opt-in local MCP server — loopback-only, token-protected; Claude Code, Cowork and others can list, read, search and add notes to your Minds
Grounded-in-N-sources chip with relevance scores and click-through to the source file
Agents are real Agents now and live within your Triplo AI
Pi Agent ships inside the installer — no runtime to install, no port to open, your Connections become its model list, your Instruction and language are its system prompt
Hermes still supported — auto-discovery of a local gateway, command palette, raw pass-through
Around the app
Resizable window — drag any edge, size remembered, dialogs keep their designed size
Backup and Restore to a single validated .zip with auto-restart
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
Version 6.x 2025–2026
6.1.0Desktop
The NEW Dictation mode is here! Hit ALT+D (Mac) or Ctrl+D (Linux/Window) and transcribe your voice anywhere (under Triplo AI Allowance)
Enhanced security measures
General bugs fixes
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
6.1.0Mobile
NEW Light Mode for Mobile
Bug fixes that were affecting Android Tablets and Apple iPads (large formats)
iOSAndroidChromebook
6.0.3Desktop
Resolved a bug that prevented line breaks from appearing in prompts
Enhanced security measures
Improved integration with custom endpoints (Connections)
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
6.0.2Desktop
YouTube Scraper is back up and running smoother than ever. Fingers crossed that everything stays smooth sailing from here on out 🤞
Added a language indicator for the current master response on the main interface
Fixed connections errors for LLM Gateway and Opper under Control Center
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
6.0.1Mobile
Fixed Chat Mode content persistency
Fixed capitalization issue in SmartPrompts
Improved stability and reliability of the Magnets feature
Added copy button to Chat Mode. Copy responses with a single click
Sorted lists of Minds, SmartPrompts, and Instructions
Refined Magnets experience (less bugs on Beta)
Removed outdated notification pop-up
iOSAndroidChromebook
6.0.0Mobile
Full Connections functionality on mobile.
New magnets feature for quicker workflows.
Quick toggle for scraping on/off.
Custom SmartPrompt list now opens directly to your selected SmartPrompt.
Improved image processing speed and accuracy.
Fixed MINDs prompt issues and search for shared SmartPrompts.
Resolved image loss bug during re-prompting.
iOSAndroidChromebook
6.0.1Desktop
Updated announcement now live with this release
Resolved bug impacting Custom SmartPrompt screen for select users
Corrected error messages overlooked in previous release
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
6.0.0Desktop
Introduced Magnets: snap Models, Instructions, Minds, and Automations directly to SmartPrompts for faster, more consistent results. 4x faster than selecting each manually. AMAZING experience for heavy users.
Automations can now be triggered via hotkeys, bypassing confirmation steps for instant execution.
Added an error message when the file is too large (10MB).
Fixed the error message for non-verified OpenAI accounts.
Reverted the behavior of clicking on the Tray. It now opens the Triplo AI User Interface again.
Improved Anthropic API integration. Better stability now.
Changed the link on the Desktop app during TRIAL to correctly point to https://www.triplo.ai/#plans
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
Version 5.x 2025
5.5.1Desktop
We Fixed That Annoying Image Bug that was affecting a portion of Triplo AI users
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
5.5.0Desktop
New Feature [BETA]: Through Control Center (https://control.triplo.ai) you can now connect Triplo AI to multiple compatible OpenAI Endpoints. This is currently available for all Desktop users in non-streaming mode. Expect improvements on this feature in future updates.
Major improvement on image processing (better image recognition and text extraction capabilities)
Fixed model indicator under History/Log
Bug fixes
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
5.0.0Mobile
UI redesign: We've standardized button shapes for a cleaner, more consistent look, fine-tuned the alignment of input boxes and chat prompts, and adjusted icon spacing so you won't accidentally tap the wrong thing.
Refined and more stable UX: Consistent placement of buttons/actions/notifications across the whole App.
Major improvement on image processing.
Performance Boosts: Fixed the lag on "thinking" models
iOSAndroidChromebook
5.4.0Desktop
Introduced the ability to select models by device in the Control Center.
Resolved issues with GPT-5 results when accessed via hotkey.
Corrected the error message for unverified OpenAI accounts using GPT-5.
Customizations (Minds, Custom SmartPrompts, etc.) are now organized alphabetically.
Fixed a bug that hindered the correct search of TAGs with uppercase letters.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
5.3.1Desktop
Resolved the GPT-5 integration issue that was hindering OpenAI from returning results to Triplo AI.
Fixed a bug on Minds that prevented content from being displayed on the original device after sharing.
Deprecated Tweak feature.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
5.3.0Desktop
Resolved the streaming issue, allowing you to follow the generation in real time.
Completely revamped the Device Management process - you can now manage licenses (multiple) and devices via our Web App at https://devices.triplo.ai/.
Introducing "Granular Sharing", allowing you to specify which devices can access your Custom SmartPrompts, Minds, Instructions, and Automations.
Enhanced the speed of Local LLMs operating on mobile devices.
Increased processing speed for Minds.
Implemented minor user interface fixes.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
5.2.1Desktop
Fixed YouTube scraping functionality
Added support for extracting captions from YouTube Shorts and live stream links
Resolved an issue where the text-to-speech icon failed to appear despite Voice Settings being enabled
Eliminated the uninstallation popup during auto updates (effective in the next update)
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
5.2.0Desktop
Resolved training issues related to fixed minds, enhancing speed, reliability, and interface.
Enhanced search results for vague queries.
Introduced a toggle to show or hide general SmartPrompts.
Awareness prompts are now concealed when Awareness mode is disabled (Context, Selection, Clipboard).
Incorporated math rendering support (LATEX).
Prevented code overflow by default and added a Wrap/Scroll button for user preference.
Improved the layout of the settings page.
Added a button to scroll to the bottom.
Implemented Home key functionality to scroll to the top and End key functionality to scroll to the bottom of the chat.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
5.1.1Desktop
Fixed a bug preventing the uninstaller from functioning correctly on Windows (damn!) when the app was open; the uninstaller now closes the app prior to proceeding.
Added the option for users to remove local settings and chat history during the uninstallation process.
Fixed a problem where auto scroll would not activate if the user prompted while not at the bottom of the chat.
Resolved an issue where auto scroll did not fully reach the bottom of the chat, preventing the display of all buttons after a response.
Addressed an issue where streaming responses would occasionally display all at once instead of gradually.
Resolved issues that caused the training process to become overloaded with incorrect tokens.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
5.1.0Desktop
Hotkeys and OOO triggers now activate Instructions, Training, Smart Prompts, and Automations wherever you initiate them.
You’ll experience a significant boost in speed and performance during AI generation.
The new Triplo AI interface now includes Japanese localization.
Scroll management during content generation has been enhanced for smoother navigation.
Fixed a bug that prevented token replenishment from displaying
Fixed an issue where long or delayed generations could override ongoing ones.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
5.0.0Desktop
Users can now leverage their Mobile Triplo AI to securely connect to local LLMs running on desktops via encrypted communication, starting with Mobile Clients v3.4.0.
Local LLM models now support Ollama and LM Studio.
Admin devices can rename other devices linked to the same key.
Minor adjustments to the user interface.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
Version 4.x 2025
4.1.3Mobile
Introduced the ability to select models by device in the Control Center.
Voice Mode now disables correctly when the OpenAI key is deactivated.
Image upload and recognition functionality restored on Mobile.
Chat mode retains previous content when switching from regular mode.
YouTube scraper feature is now fully functional on Mobile.
iOSAndroidChromebook
4.1.2Mobile
Voice Mode now disables correctly when the OpenAI key is deactivated.
Image upload and recognition functionality restored on Mobile.
Chat mode retains previous content when switching from regular mode.
YouTube scraper feature is now fully functional on Mobile.
iOSAndroidChromebook
4.1.1Mobile
Integrated GPT-5
Mobile app now connects with the Triplo AI Control Center
Implemented minor UX/UI enhancements
iOSAndroidChromebook
4.1.0Mobile
Added the ability to process image and text files (PNG, JPG, PDF, TXT, DOCs, etc.) on mobile devices
Resolved problems with scraping YouTube links.
Included Brazilian Portuguese (❤️) in the list of YouTube languages.
Resolved the logging issue for prompts containing scraped content in the history.
Preparation for the Device Management Dashboard (to be launched)
iOSAndroidChromebook
4.0.0Mobile
Significantly enhanced user experience (faster load times, fewer validations, reduced risk of accidental app closure)
Launched a new Features Hub section
Boosted the loading speed of SmartPrompts at startup
Revamped the design of the Settings section
Improved user interface for the Voice Mode experience
Models used included in history
Added token consumption information to history
Improved recall in History generation
Corrected font color in Markdown tables
Enhanced readability with larger fonts
Refined device management logic during updates
iOSAndroidChromebook
4.1.0Desktop
Incorporated OpenAI o1 and o3 models under BYOK.
Ollama model endpoints are now compatible with OpenAI [BETA].
Introduced a Spider toggle in the UI footer for inline scraping.
Added hotkeys: Control+Shift+S to toggle inline scraping and Control+Shift+C to switch to chat mode.
Enabled reasoning display for OpenRouter models that support it.
Enhanced accuracy in token usage measurement, now sourced directly from the service provider.
Improved display and error handling for slower models, including Deepseek R1 and other cognitive models.
Resolved issues with the listing of Ollama models.
Fixed Mac compatibility issues related to trigger and hotkey configurations.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
4.0.0Desktop
Triplo AI can now access and process Local Models via Ollama!
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
Version 3.x 2024–2025
3.4.1Mobile
Local LLM models are now available both with and without BYOK activated.
Resolved the frustrating bug that caused Triplo AI to forget the last selected model.
iOSAndroidChromebook
3.4.0Mobile
Triplo AI Mobile can now connect to local LLMs operating on Desktop Clients (from 5.0.0 on).
Enhanced UI/UX with a fresh design.
Improved speed thanks to a new AI loader.
Fixed the bug related to sharing with Triplo AI.
iOSAndroidChromebook
3.3.2Mobile
Token counting error has been fixed.
Markdown rendering is now supported in both single prompt and chat modes.
Minor bug fixes.
iOSAndroidChromebook
3.3.1Mobile
A correction has been implemented to resolve an issue that was erroneously causing token exhaustion.
iOSAndroidChromebook
3.3.0Mobile
Major performance improvement
Added switches for BYOK
Bug fixes and general improvements
iOSAndroidChromebook
3.5.0Desktop
Localized the interface to German (DE).
Enhanced the device association reset process.
Introduced tooltips for features in the Settings/Switches menu.
Added direct links to Help and Documentation for key features.
Implemented a confirmation prompt for deleting history/logs.
Included token estimation for attachments during prompts.
Masked API keys and license details to enhance privacy.
Added a reset link in the Activation/Device Management screens.
Included model names in history/logs.
Introduced a "Shift+Backspace" toggle switch in settings.
Added voice speed control in the Voice settings.
Expanded the Voice feature (TTS) with new voice options.
Implemented a "Re-Prompt" option in the main input interface.
Resolved minor UI issues.
Made security enhancements.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
3.2.2Mobile
Fixed Anthropic Integration
Fixed Long Results Bug
Security Improvements
iOSAndroidChromebook
3.2.1Mobile
Models list can now be updated without requiring a new release
Resolved an issue with the Minds on Mobile
iOSAndroidChromebook
3.4.1Desktop
Three hotkeys are back!!!
Increased limits for Hotkeys: Early adopters can configure up to 50 per device, Soft Plan users can set 5 per device, Medium Plan users can set 15 per device, Pro account holders can set 50 per device, and Free Plan users are limited to 1 per device.
Significant speed enhancement for both Mac and Windows.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
3.4.0Desktop
You can now incorporate images [BETA] into your prompts by pasting them from the clipboard or attaching them directly.
Introducing Triplo AI Hotkeys, which can process pre-defined or inline SmartPrompts on highlighted text and automatically replace the original content.
Users now have the option to allow other devices to "Edit" their shared SmartPrompts, Automations, Instructions and Minds.
Minds can now be trained using multiple URLs simultaneously.
SmartPrompts can be organized and sorted by tags, complete with search and filtering options.
Various minor UI improvements have been implemented.
A bug that prevented file attachments when the UI wasn't pinned has been fixed.
The models list can now be updated without requiring a new release.
Triplo AI now allows you access to 60 models. Details [here](https://triplo.ai/docs/models).
🔥🔥🔥
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
3.2.0Mobile
Added GPT-4o-mini, Jamba 1.5 Mini 256k, DBRX 132B Instruct 32k, Microsoft WizardLM-2 8x22B 65k, Mistral Nemo 128k, NVIDIA Nemotron 70B 131k, and Qwen2-VL 7B Instruct 32k under Allowance.
Users can now use their Anthropic key to access premium models, including Claude Haiku 3.5
Minor Bug Fixes
iOSAndroidChromebook
3.2.0Desktop
Introduced the FILES feature on the Training page and Prompts.
Updated model lineup, now including: AI21 Jamba 1.5 Mini 256k, DBRX 132B Instruct 32k, Microsoft WizardLM-2 8x22B 65k, Mistral Nemo 128k, NVIDIA Nemotron 70B 131k, and Qwen2-VL 7B Instruct 32k.
Removed outdated models: Mistral Open Orca 7B 8k, PPLX Online 7B 4k, and Yi 6B 4k.
Enhanced UI performance on the Training page.
Implemented minor UI/UX improvements.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
3.0.1Desktop
Allowance tokens are now standardized at 2,000,000 tokens per device per month across all 15+ available models, including the GPT-4o Mini.
Resolved several bugs introduced in the latest Windows version, such as the black UI, single-word prompt issues, and JavaScript errors.
Fixed a problem that prevented users from copying results from the UI on Windows 11.
A new hotkey has been added to restore the UI display; simply press CTRL+SHIFT+P while Triplo AI is in focus.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
3.0.0Desktop
Triplo AI is now equipped for multiple languages, featuring localized user interfaces in Arabic and Brazilian Portuguese.
We have enhanced the multilingual experience by adding a language selector in the main UI, allowing users to prompt models to respond in their preferred language.
Resolved an issue in the automation creation form where the 'suggest payload' button was non-functional for non-string parameters.
Security updates have also been improved for better protection.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
3.0.1Mobile
Added multiple new models under OpenRouter: Airoboros L 2 70B, Code Llama 34B Instruct, Goliath 120B, Mistral Open Orca 7B, Mythalion 13B, MythoMist, Neural Chat 7B, Noromaid 20B, Nous Hermes 13B, Open Hermes 2.5, Phind Code Llama 34B, PPLX 7B, Psyfighter v2 13B, Synthia 70B, Yi 34B, Yi 6B.
Multiple small fixes for better UX using Voice Mode on Triplo AI Mobile
Fixed bug that unexpectedly triggers inline SmartPrompts.
iOSAndroidChromebook
3.0.0Mobile
Release of Voice Features [Beta]
Changed Mixtral model under Allowance to Dolphin version
New Models added under Open Router keys
Added Multilingual Youtube Scraping features to Mobile
Refined UI (better icon placement, fonts and other improvements)
Improved allowance models selection
iOSAndroidChromebook
Version 2.x 2023–2024
2.9.0Desktop
Included GPT-4o Mini under the allowance of 1 million tokens per month per device and API.
Added an Anthropic API key so users can access Claude models directly.
Added Claude 3.5 (OpenRouter/Anthropic), LLaMA 3.1 (8B & 70B under allowance/API), and 405B (API).
Cleaned up deprecated OpenRouter models under allowance/API (Nous Hermes Llama2 70B, Zephyr 7B, Synthia 70B).
Added inline scraping and API key switches under settings.
Significantly improved speed on Macs.
Improved UX: On errors, pressing "ESC" returns you to prompting or halts the current generation; "SHIFT+BACKSPACE/DEL" clears the previous results.
Fixed issues with agents and the unexpected "vanishing" UI in Windows.
Fixed automation (webhooks) under trial.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.8.1Desktop
Fixed UI clearing after the response when using Openrouter models.
Fixed the settings persistance for Chat Mode, Active Mind, and Active Custom Instruction.
Added the following models to OpenRouter (API): Gemini Pro 1.5 and Gemini Flash 1.5
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.8.0Desktop
Included OpenAI's new GPT-4o model.
Implemented security updates on the backend.
Minor bug fixes.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.7.1Desktop
Fixed Settings "Save State".
Fixed Google Gemini cutting off results.
Adjusted Google models tokenization (different from other models).
Google Gemini Pro under allowance is now paired with Open AI's GPT 3.5 not the other OpenRouter models.
Fixed Chat history markdown crash for some markdown tables.
Added Llama3 8B to the Allowance (1/4 tokens cost model).
Added Llama3 70b to the API Key usage.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.7.0Desktop
Expanded allowance for plans (500,000 GPT 3.5 and 2 million Mixtral/OpenRouter tokens per device per month)
Added multiple OpenRouter models to the allowance. Now you can use the following models: Dolphin Mixtral 8x7B 32k, Google Gemini Pro 131k, Meta Llama-2 13B 4k, Mistral Open Orca 7B 8k, Mistral Tiny 32k, Mistral-7B 8k, MythoMax 13B 4k, MythoMax 13B 8k, MythoMist 7B 32k, Nous Llama-2 13B 4k, Open Hermes 2.5 7B 4k, OpenChat 3.5 7B 8k, PPLX Online 7B 4k, Toppy M 7B 32k, Yi 6B 4k, Zephyr-7B 4k
Introduction of new models: Claude 3 200k Haiku/Sonnet/Opus, Mixtral 8x22B 65k, and the new GPT 4 Turbo
Introduction of a new lifetime FREE plan for Triplo AI
Clearer pop-ups and error messages on disabled/limited features
Bug fixes and minor UI improvements
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.6.0Desktop
Triplo AI UI can now be resized
The Status Dot (visual notification of processing status) is back!!!
You can now use TAB to select your SmartPrompts, Awareness commands etc
Added new models from OpenRouter: Airoboros L 2 70B, Code Llama 34B Instruct, Goliath 120B, Mistral Open Orca 7B, Mythalion 13B, MythoMist, Neural Chat 7B, Noromaid 20B, Nous Hermes 13B, Open Hermes 2.5, Phind Code Llama 34B, PPLX 7B, Psyfighter v2 13B, Synthia 70B, Yi 34B, Yi 6B [DONE]
GPTs 3 and 4 models updated to their latest versions (0125)
Minor UI improvements: fixed double Enter bug on Voice mode, added an alert for Automations under OpenRouter, corrected bug related to RTL languages.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows [MANUAL INSTALLATION]
2.5.6Desktop
Corrects bugs introduced on the latest version 2.5.5) related to the initial use of Mixtral/GPT allowance Models.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.5.5Desktop
Added voice features (Text to speech and speech to text) with keyboard shortcuts
Replaced mixtral allowance from Mixtral 8x7B Instruct to Dolphin Mixtral 8x7B 32k
Added Scrapemark SmartPrompt and inline markdown scrape with #url (Ex. #https://triplo.ai)
Added the following models to OpenRouter: Perplexity 70B 4k, Dolphin Mixtral 8x7B 32k, OpenChat 3.5 7B 8k, MythoMax 13B 4k, MythoMax 13B 8k, Mistral Tiny 32k, Mistral Medium 32k, ReMM SLERP 13B 6k, Izlv 70B 4k.
Fixed a wrong error message in inline scrape where it would show invalid URL instead of 'couldn't scrape this site' error
Tweaking feature is now out of BETA
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.5.4Desktop
Minds can now be shared between devices (Desktop and Mobile)
Added scrapehtml normal prompt and inline with @url (RAW HTML Scraping)
Changed sync approach (a device can now see synced data from others even if it's sync is off)
Added auto translate to YouTube scrape (AMAZING!)
Dark/light mode control under settings
UX Improvement - "esc" on secondary screens returns to the main UI
Fixed app crash in history page when some chats have an error in inline scraping
Fixed an issue on Windows and macOS where awareness did not work with Ctrl+A in some cases
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.2.7Mobile
Minds can now be used on Mobile
Added Mixtral model on the Allowance Tokens
Minor bug/UI fixes
iOSAndroidChromebook
2.5.3Desktop
Released multi-URL scraping (you can now scrape multiple pages at once inline)
Instructions can now be significantly longer (5000 chars)
Improved Allowance Tokens process
Added a debug info extractor under settings for better support
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.5.2Desktop
Allowance has been updated to include Model Mixtral 8x7b 32k, allowing for up to 1.5 Million tokens per month per device
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.5.1Desktop
Major performance fixes and improvements
Added Mixtral 4x7B Instruct to OpenRouter models
Improved error handling to inline scrape (Triplo AI warns of scraping problems even on inline prompting)
Added tokens of each History entry
Added PUT on Automation settings
Bug fixes on Automation (confirmation sometimes caused the app to crash) and UI bug in History page if some titles were too long
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.2.6Mobile
Implementing OpenRouter
Improve device management functionality
Improve chat title functionality
Security improvements
Bug fixes
iOSAndroidChromebook
2.5.0Desktop
Expanded Training Minds to 200k tokens
Reduced training tokens costs to 1/5th the chat/prompting costs
Training Feature is not under Beta anymore
Added ability to disable confirmation of Automations under settings
Automation Feature is not under Beta anymore
Added ability to rename chats, AI rename chats, and automatically naming chats on creation
Added copy button in chat history
Fixed a but were some new chats were not shown in history due to a refresh issue
Improved license validation and device management UI/UX
Improved error messages and error handling
Fixed a bug of stop generating content
Added 'Auto Select' for GPT3.6 16k and GPT4 128k under OpenAI/OpenRouter APIs
Added Tobby M 7B 32k under OpenRouter Models
Faster OpenAI and OpenRouter keys validation
Improved Allowance Keys security
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.2.5Mobile
Now mobile App can act as Admin Device
Create/Edit/Sync Custom SmartPrompts
Improvements on Instructions Syncing
iOSAndroidChromebook
2.4.8Desktop
Allowance tokens now are GPT3.5 16k
Added OpenRouter models: Hugging Face Zephyr 7B (beta), Hermes Llama2 13B (beta) and Hermes Llama2 70B (beta)
Added Admin approval for new devices under Devices Management
Added button to delete all chats
Added confirmation screen on delete (chat history, custom prompts, instructions, automations, and training)
Added “~” as a hotkey option
Security improvements
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.4.7Desktop
Code refactoring/cleaning. Preparation for future developments.
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.4.6Desktop
Security fix
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.4.5Desktop
Speed improvements
Fixes for multiple keyboards/languages
Security improvements and bug fixes
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.2.4Mobile
Bug fixes
iOSAndroidChromebook
2.4.4Desktop
Hotkeys Fix/Revision
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.4.3Desktop
Server Upgrade
Hotkey scheme change (2 keys)
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.4.2Desktop
Server Upgrade
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.4.1Desktop
New Device Management dashboard (with sync settings)
Fixed an issue with Youtube scraping
Improved control of Automation (confirmation routine)
Automatic text alignment for RTL/LTR
Added Mistral 7B Instruct model to OpenRouter
Better UX for Training setting
Stopped showing window on startup
Fixed autoscroll bug (it didn’t scroll all the way to the buttons)
Disabled automations on OpenRouter (not supported)
Added more choices in hotkey options
Performance improvements
General bug fixing
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.2.1Mobile
Server Upgrade
iOSAndroidChromebook
2.2.0Mobile
New Website/YouTube scraping (Scrape/Inline)
Improved Awareness Support (apps with no available "share" function)
Added Custom SmartPrompts sync
Fixed Chat Mode on iOS
Fixed iOS notifications issue
Improved loading speed on the first login
Changed the history titling prompt
Added Welcome screen
iOSAndroidChromebook
2.1.4Mobile
Bug fix iOS old iPhone models
iOS
2.1.3Mobile
Custom SmartPrompts syncing
Bug fixes
iOSAndroidChromebook
2.1.2Mobile
New Awareness Mode for Mobiles
Fixed Chat history order
Improved Chat persistency
Many UX improvements (less clicks/touches)
Bug fixes and Code improvements
iOSAndroidChromebook
2.1.0Mobile
Added Instructions feature
Much better/polished UI
Several bug fixes
iOSAndroidChromebook
2.4.0Desktop
Integration of OpenRouter API [BETA]
Added Always On mode
Fixed License Revoking Bug
Fixed Windows Scaling/Multi-Monitor Bug
Linux (AMD/Intel)Mac (Intel/Silicon)Windows
2.0.0Mobile
Unified version Android/iPhone
New polished User Interface
Added new Chat mode support
Added new History/Logging capabilities
Added the GPT 3.5 4k tokens for licensed users
Added model selection
Bugs fixes and Improved stability
iOSAndroidChromebook
2.3.6Desktop
New Tweaking [BETA] feature (Creativity/Diversity settings)
Add option to set Youtube language for scraping under Settings