MakuluLinux Max Beta 2 Release Notes :
MakuluLinux Max is the first of its kind, an AI Integrated Operating system, Designed with Debian compatibility and running on the Cinnamon Backend Framework, Utilizing Multiple Desktop Layout features and offering a lot of flexibility in terms of theming and customizability options. It comes out of the box with a variety of options to style and theme and customize the OS. What makes the operating system unique is not only all the amazing theming and appearance features, instead its the Artificial Integration options that really stand out. You can communicate with AI using Voice, Widgets, Text, Web and Terminal interface, making it remarkably easy for anyone to communicate in their own preferred comfortable method, Not only to communicate with the AI, but you can control most of you computer just using your voice. To top that, the operating system also comes with a whole bunch of custom written AI applications, created by MakuluLinux to give that extra unique experience. This is really one of those very rare, very different releases that doesn’t come around very often. The future is changing quickly, and MakuluLinux has taken a bold step with Max to stay ahead and make sure its user base are out there on the front lines enjoying all that the new technology has to offer. Let’s take a look at some stand out features and some greater detail of what MakuluLinux Max is all about.
Some Stand out features (in no particular order) :
Integrated A.I Voice ( Electra ) :
MakuluLinux has been spending months creating its own AI model called Electra. some clever coding and Combining it with Palm2/Google and Bing to bring you the Ultimate integrated AI, Running 50 processes, capable of handling an average of 400 queries per minute ( on the current hardware ). Capabilities include communicating in natural language, answering questions in a comprehensive and informative way, generating different creative text formats of text content, Collaborating on tasks like code generation, generating code snippets, debugging, explain code, writing jokes, music, social media content, blogging, Generate summaries of web pages, researching and communicating using voice input, etc. etc. the list goes on and on and the AI learns and grows more everyday, so does its capabilities. The AI is very smart and typically replies on average in around 7 seconds, You can ask it anything. due to clever coding, it also has web access, so it can do web searches, answer questions about web content, debate on current affairs and even tell you what’s new on Netflix. She is very good at writing content, and can even generate you a bed time story.
Unlike a lot of other AI’s, Electra has an excellent memory, You can debate with her for hours on end and she will remember every word. She uses a “Conversation ID” to keep track of all conversations. The conversation ID gets reset if you relog, refresh the page, exit the browser/widget/terminal interface.
You can communicate using the Built in Voice integration, This operates much the same way as siri and google does on mobile handsets. Instead of saying hey siri or hey google, You simply say “Hey Electra” to wake her up, and once she is awake you can converse with her using voice. the system has been designed and optimized so well that you have an average of 7 seconds communication delay. that means you get a reply in under 10 seconds.
You can also communicate with Electra using the Terminal Interface that you launch from the desktop, here you can use either text or voice to communicate while seeing the responses typed out back to you which makes it nice if you want to copy/paste for storage or sharing.
Other options include a Desktop Widget that you can run from the panel that allows you to communicate with the AI using text in a mini browser environment, the widget is resizable and draggable. You can also communicate with Electra via our Web interface and our desktop chat client which you can launch from the main menu.
Electra Also has an “Information Mode”, and this mode is accessible only via the Web/Android interface. This mode is specifically designed to give quick results with web links to reference data. If you are researching a topic and you need some information quickly and verifiable links, then information mode is just what you need. response time in Information mode is typically 1 to 3 seconds.
The Third and Final part mode of the Voice system is “System control”, You can control your computer using just your Voice. There are several key areas that you can control on your system simply using voice commands, these are the following :
- Opening / Closing / Minimizing / Maximizing / Restoring Applications.
- Running System Commands.
- Dictating.
- Controlling / Navigating Chromium using Voice.
- Web Search using voice ( no browser ).
- Mouse control ( Mouse Movement, Mouse Clicks ).
- Custom Application control.
- Widget Control.
- Wallpaper control ( setting wallpapers using voice ).
- Layout Control, Changing between Layouts.
- System control, using voice to run system updates, restart shell, reboot computer etc. etc.
To use AI Voice simply say “Hey Electra”. this will wake up the system.
You can then communicate with the AI in voice Saying “Electra [your question]”.
Example : “Electra what is the latest news headlines”.
Speak Clearly, We use the Vosk Voice recognition system, It is one of the Best Free options, and it works generally well if you have a clear accent. if you don’t, try using the various text methods available to communicate with electra instead.
You can find all the available “Voice Commands” by opening “Makulu-Voice” shortcut in the main menu.
Not all voice commands are available in the regular free build of Max. Some voice features are only available in the Pro build.
Virtual Cam :
MakuluLinux has Designed from scratch a Virtual Camera. This is not simply any camera application, This is specifically focused on being a “One Click” option where you click the shortcut in the menu and the camera app opens and works out of the box. No scripts to run, no commands to issue, no guides to follow, “click the shortcut, and the app starts running out of the box”. Once the Virtual cam is launched, it will be available to use in any third party conference application like Zoom, or Web browsers that have streaming features. This cam is perfect for live video conferencing, streaming, teaching online or any streaming situation where others see you live onscreen. What makes this Camera app so special is that it supports some really nice features out of the box :
- Blurring the Background – it separates the background and the subject, allowing you to blur the background using a slider. Like a greenscreen effect, without the green screen.
- Fading the Background to white – If you prefer to remove the background instead of blurring it, it separates the background and the subject, allowing you to fade the background using a slider. Again Like a green screen effect, without the green screen.
- Skin Smoothing – You can use the Skin smoothing slider to simulate a “De Aging” effect, remove wrinkles or beard stubble and just give it that slight beautifying effect.
- Various Lighting controls – “Brightness / Contrast / Gamma / Saturation / Sharpen” that help with making you look great onscreen.
- Adding a Fake Background Image or video – Users can select to add a fake image or video for their background without needing to use a greenscreen.
Makulu Virtual Cam is the perfect tool for anyone doing any kind of live streaming or video conferencing while wanting to Blur their background. Essentially this lets you sit at home and not have to worry about people seeing your room for whatever reasons, maybe because it is messy, maybe you don’t want something classified to be seen by prying eyes. Whatever the reason, you can now simply Blur or fade the background easily with the virtual cam. Video conferencing and controlling what others see has never been easier.
AI – Text to Image Generator :
MakuluLinux has build a custom A.I – Text to Image generator. This application lets you generate an image from a word or phrase. You can simply type for example “Hippo wearing cool sunglasses and a hat” and hit generate, the generator then generates 9 images of a “Hippo wearing cool sunglasses and a hat”. whatever you type, it will generate. and it does so fairly quickly, 9 images in around 60 seconds, that is 6.6 seconds per image. It also supports various image types. This little tool is perfect if you need an image for a spreadsheet, essay, blog, presentation, web page, or any situation where you quickly need an image. and the best part is it generates what you ask it to, even if you say “astronaut monkey”, that’s what it will generate. Like all our other applications, it is also a one click app. run the shortcut from the menu and that’s it, no scripts to run, no commands to issue, no guides you need to read.
Auto-Gpt ( AI Agents ):
We have made a simple GUI that lets you run Auto-Gpt out of the box with very little effort, just enter your openai key, click and button and off you go. But what is Auto-Gpt ?
Autonomous A.I agents is the latest trend, where users can ask the A.I to perform a specific task and it goes to work to achieve that goal, and it never stops until the goal is achieved. The A.I will look and analyze the task you specified, make a list of mini tasks it needs to follow to achieve the goal, then implement it, we refer to this A.I as an “Agent”. It also has a self checking process where it evaluates it own progress and adjusts the tasks as needed. It can, and often does assign additional agents assist with the task. Some examples of things you can ask it to do :
- Find products online with specifications.
- Research something.
- write something ( a book, code, a research paper, essay etc. etc. )
- create an application
- Pretty much anything you can think of…
I once asked it “look at the stock market, find stocks that normally trend upwards, but are currently trending down, also find the best website with good ratings to trade stocks”. I then watched it go to work, it literally started scanning stock websites, looking at specific companies, going through ballet sheets, back dating investments made by companies, it even looked for social media and announcements companies had made that could influence its stock. it was amazing watching the process.
Another example was when i asked it to write a book and i gave it a titles and some context (children living on a colony on the moon ), I specified the type of book as a “novel”. I watched this A.I teach itself to write, it literally went to websites that teach you to write, writing styles etc. etc. and then it started researching the moon, it also researched other books and novels on the moon, I was quite amazed at how efficient the A.I was at accomplishing tasks i gave it.
Story Teller ( AI Content Writer ):
Content writing in this digital age is pretty much a standard thing for most people. People write content in their daily lives, whether you are writing a blog or a presentation for work, writing is a given in today’s times. Story Teller is the perfect application for content writing, Utilizing the power of A.I to quickly and efficiently give you top quality content. Simply select the content type and either give it keywords or a url and it will do the rest. You can for example select “Blog and type : The dangers of second hand smoke” and hit enter and it will write a full fledged blog post on the dangers of second hand smoke. Its truly amazing. it supports the following forms of content :
- Advertisement
- Blog posts
- Essay writing
- Bed Time Stories
- Facebook Posts
- Jokes
- Public speeches
- Short Stories
- Summarizing
- Twitter Tweets ( with 280 character limit )
Story Teller uses Electra to generate content and She is very good at it. If you are a content writer this application is a dream come true, you can now simply select your content type, feed it some keywords or an url ( or both ) and hit submit and watch it work its magic, saving you hours and hours of work, in mere seconds it generates you premium content for free !
Electra AI Terminal Interface :
Electra Terminal Mode let’s you communicate with the AI using various modes directly from the Terminal, You can access Electra using Voice or Text and some people are just more comfortable in a terminal. Electra has web access, so you can ask her anything about anything, even current affairs, she has an excellent memory so you can really have full fledged conversations and debates with her that can last hours… However she also has Read Mode : Read mode let’s you type words or phrases and the A.I reads these back in AI Voice, this is perfect if you want something recorded in AI voice. simply start your voice recorder, enter read mode, type a phrase and record the A.I saying the words. now you can attach this voice to anything… Electra Terminal also has a multi Line tool, since you can only type 1 line at a time in the terminal we designed a “multi line” interface that allows you to paste or type as you normally would in any input box with no line limitations and send to the AI. perfect if you want to paste large sections of code or text. Electra Terminal is a very well designed interface for communicating with the AI.
Electra can also be accessed from the panel Widget or from the menu, both these shortcuts open a desktop GUI widget client that can be used to communicate directly with the AI via a web like interface. It uses a bubble chat scroll type chat that makes it nice and comfortable to use.
ChatGPT and Other AI interfaces :
MakuluLinux Max provides Desktop GUI Clients for Chatgpt / Bard / Bing and Other A.I systems, allowing users to easily connect to their Favorite AI without needing to leave their desktop. You can now easily log into your chatgpt account directly from the desktop client, it fully supports all the chatgpt features including the Plus features and chat history.
Chromium Browser integration :
We have also added various Web extensions to access AI features from within the Chromium Browser as well as some well placed Favorite tiles to quickly access your favorite search engine or AI generator tool.
Widgets :
MakuluLinux Max has a Flexible Widget system that lets uses run “Widgets”, these widgets run in their own containers, like little browser windows that are draggable & resizable . Currently we support 24 Widgets, Widgets range from AI to Maps, to Social media, to Movie / Tv Series. The Widgets system lets users access these applications/websites with total privacy, there is no history stored, no cookie, no cache. Complete privacy. Perfect for anyone who prefers their privacy. The other major advantage about widgets is that they sit in the panel and can be popped open with a single click whenever a user needs to access them. The widget system is forever expanding. We will continue to make new widgets in the future.
Desktop Manager :
The MakuluLinux Max Desktop Manager let’s you customize and theme your Desktop. You can select from System Layouts, Constructing tools, Special Effects, Cursors, Icon Sets, wallpaper management and Theme colors. All changes are almost instant, no scripts required, no commands, no reboot required. Simply click what you want to change and its done.
Desktop Theming features :
- 155 wallpapers
- 20 cursors
- 10 Icon sets ( each set sports multiple color variations )
- 12 Light themes and 7 Dark themes for a total of 19 theme color variations.
- 3 Layouts ( Dock layout and 2 Panel Layouts )
Software Center :
MakuluLinux uses the Default Gnome Software Center ( With some tweaks to improve performance ), The Software Center in Makulu Shift fully supports both Flathub and Snaps, Users can install from a selection of Thousands & Thousands of great software and games with a few easy clicks.
Gaming :
MakuluLinux Max comes with Steam pre installed, it’s so easy to game on Max. Simply open steam, login and play. Any additional gaming platforms you may want to install you can grab from the Software center. MakuluLinux Max is game ready.
Constructor Tool:
Constructor Tool allows users to recompile your operating system back into an ISO, so you can reinstall it on other computers. It takes a system snapshot of your computer, after you have made customizations and changes. and this snapshot is carried over into the ISO it creates. it is the perfect tool for re-spinning Makulu into your own distro, or simply making a backup to re install at a later date.
Base Specifications of MakuluLinux Max :
- Code Name : Max
- Base : MakuluLinux Hybrid Base
- Life Cycle : Semi Rolling
- Main Repository : MakuluLinux
- Secondary Repository : Debian Testing
- Desktop Environment : Gnome Framework.
- Kernel : 6.0.0.5 ( upgradeable )
- Architecture : 64Bit with 32Bit support ( 64Bit CPU is required )
- Package system : Debian
- 3 Unique Desktop layouts.
- 1 Click Automated Switch between Layouts.
- 155 Wallpapers.
- 10 Icon sets
- 20 cursors
- 19 GTK Theme Variations
- Huge Driver Support
- Some Software Pre-installed and Pre-configured out of the box.
- Required Disk Space : 15Gb+ minimum / 20GB+ more recommended
- Required Ram : at least 2GB ( 4GB+ recommended )
- Live username / password : makulu / makulu
- Release Date : June 2023
- Download size : 3.20xxxGB
Designed for Speed and stability
We spent a Lot of time optimizing MakuluLinux Max. While our main objective was not the lowest ram consumption, Speed, Snappiness and Stability however was high up on our list of goals. You will feel right out of the Box how fast Max feels, it runs so well that you may just forget it is running on a Gnome Framework. It certainly doesn’t feel like it, and it feels very very stable, When we got to that point we knew we had just the right balance. I’m quite sure many reviewers may end up commenting on how snappy the system feels considering what’s all under the hood.
Information About the AI Base System
In our Very Early builds of Max we initially implemented a Openai AI model, however this ended up being quite problematic for us for a number of reasons. firstly it required users to register and generate API keys, and these can be costly to the end user, not to mention a nightmare to setup for people who are not technically minded secondly it had no internet access so it ended up just being a clever chatbot with somewhat disappointing results. thirdly it had a terrible memory, it would constantly forget conversations… We quickly abandoned that idea.
Instead we opted to build our own, We wanted something that was free to use, but also very capable to do any daily task someone may require from AI. With a LOT OF RESEARCH, clever coding, a lot of trial and error and frustration and patience, that is exactly what we have achieved, We have surpassed any of our original goals and expectations.
Currently the AI system runs on a variety of Models, namely Llama/Palm2/Bing, with some clever coding we have taken this combination to provide you an excellent AI experience, one that can do virtually anything required, from being a friendly chatbot, to coding, to generating content, to doing research, to discussing or reporting current affairs. whatever you need, it can provide, and at the same level as GPT4, in some regards even better due to its web access. In fact, many of the AI applications we use now in Max were in part created with the assistance of Electra. the Virtual Cam, Image Generator, Story Teller, Widgets System… Electra had a big hand in development of each of those Tools. So I speak from personal experience when I say, It is very capable. It leans every day and it will continue to grow over time. AND, it has an Excellent memory, each conversation is tracked with an ID, ensuring continued un-interrupted conversion, this was also very important to us.
Important Warnings :
We have built a Great OS here, But the AI is considered both in “Beta” and “Experimental”. We have spent a lot of time and delayed the release many times making sure the OS + AI runs well and stable, and while we don’t expect any major issues, There are some concerns about the AI :
- We have no idea how many people will download and what strain it will put on the server load on our AI system as it has never been tested with Mass Use.
We do not know how many people will download this release and thus we cannot speculate the AI server load or how it will handle that load. We have tried to prepare for this. however there is always this fact :
While we offer many of our AI integration features in our Free build so that everyone can enjoy the fruits of our labor, there is NO guarantee that we always keep the AI features in the free build, if we find the server load too much to bare, we may at any point in time disable some AI features on our free Builds.
The Good news ? – We will of course keep these features running on the Pro builds. So keep this in mind when choosing which build to use when running MakuluLinux Max.
Final words :
Please remember, this is a Beta 1 release, not a final build. it may feel like a final build because we spent such a long time on this Beta and its running so stable, and can be used safely as a daily runner. Any issues you encounter please report them in our bug reporting section.
Also Remember this is a FREE project, We do not get paid for making these releases, it is very important for community to support the project if you are happy with our work. Community support will be used to pay the running costs which are not exactly cheap…
If you like our work, support us by upgrading to the Pro build or donating to the project via the donation links above. Alternatively you can become a patreon member and get perks while supporting us.
We hope you enjoy this release, We really enjoyed making it, We wanted to make something truly Unique in every sense of the word, and this was the result.