OS-Copilot: Your Personal AI Agent with Semantic Memory - Rabbit R1 Alternative That is FREE!

with the rise of rabbit R1 many of us have been Scavenging on the web to find alternative personal autonomous AI agent assistance I've covered a few on this channel that are completely for free to access such as mulon as well as hyperight well today I would like to introduce another personal autonomous AI assistant that takes control of your OS introducing OS co-pilot it's a project working towards a generalist computer agent with self-improvement now unlike the other OS AI assistants I've covered OS co-pilot is an ideal autonomous functioning agent that can interact with the computer on challenging tasks and self-improve for future tasks for having this semantic memory feature sorry for being repetitive but this month we had insane Partnerships with big companies giving out subscriptions to AI tools completely for free these are tools that will streamline your business's growth and improve your effic efficiency just being a patreon this past month you were given access to six paid subscriptions completely for free not only do you access these subcriptions but you gain the ability for Consulting networking collaborating with the community as well as with myself you get access to daily AI news resources giveaways and so much more if you're interested check out the patreon link in the description below to gain access to these benefits now just take a look at this demo video which is showcasing the use case of OS copala now in this demo we can see that there is a prompt being given to OS copala to play any random music and the copala executes this command by opening up any streaming service that is locally prevalent so in this case it's opening up Apple music to play that random music and it starts playing it and this is just something that is quite generic it's an example that is just showcasing the very basic use case of os copol it there's many other examples and this is something that we're going to take a look at throughout today's video now we're also going to be be exploring this new personal autonomous a system further in detail we're going to showus how you can install it and going over the semantic memory feature as this is something that is quite unique in comparison to many of these other autonomous AI systems so with that thought guys stay tuned and let's get straight into the video if you would like to book a one-on-one with me where you can access my Consulting Services where I can help you grow your business or basically give you a lot of different types of solutions with AI definitely take a look at the calend Link in the description below hey what is up guys welcome back to another YouTube video at the world of AI in today's video we're going to be taking a look at OS copilot this is a new framework aimed to develop generalist computer agents that are capable of interacting with various elements within your operating system this includes the web code terminals yeah have files multimedia and thirdparty applications it's introducing Friday which is is a self-improving embodied agent and it's something that is designed to automate General computer tasks Friday has demonstrated strong generalization to unseen applications and it's something that even outperforms previous methods by 35% this is off of a g Benchmark now this framework also showcases evidence of Friday's ability to control and self-improve on tasks like Excel and PowerPoint with minimal supervision it's something like semantic memory where it's able to have this reoccurring memory to improve on itself so that future Generations are better just take a look at this example over here which is showcasing what you can actually do in various types of elements in this case you're able to enter into the focus mode where you can adjust your work layout and theme you can even have it play music in this case it's calculating and filling out the spreadsheet with a basic prompt you're able able to even create a web page and it can do so by generating the files that are required it can access this on your own operating system and compile it and build your web page for you so before we even get into the nitty-gritty I'm going to be showcasing how you can quickly get started with this it's super simple you want to copy this command over here but before we even get to that there's a couple prerequisites that you will need to fulfill beforehand you're going to need to make sure that you have git which is an application that lets you clone any repository on GitHub you need to make sure that you have python as well as a code editor such as Vis Studio code once you have those three things installed you also need to make sure that you have condos to actually set up your python environment once that is done you can then move forward by installing this so it's super easy what you want to do is open up command prompt once you have done that you can then copy and paste that command that we just recently copied which is get clone the repository link and then click enter this will start cloning this repository and it will take a couple seconds and once it has done we can then move forward with the next command and that is actually creating the environment for Python and it's super simple you want to copy condic create and this command and then you can move forward with this so if it stalls a little bit you can click enter and it will proceed forward with installation and once this is finished you can then copy and paste this command into your command prompt okay let me copy it again so once that is done I can then click enter this will start creating the environment and once it has finished creating the environment we can then activate it so that we can operate whatever commands within that environment so once it has finished doing so I'll be right back now once it has finished creating your environment you can activate it with this command which is right over here once you have done that you can install the dependencies so what you want to do is go into the Friday directory and it's super simple you can just type in Friday CD Friday sorry and then click enter once you're inside the directory you can then install the requirements this will take a couple of minutes so don't worry about it you'll let it sit for a little bit and install and once that's finished installing we can then move forward with installation so in the meantime what you can do is set up your open AI API key and this is by going to the open AI API dashboard collect your API key and then we're going to be inputting it into the config.js file so make sure you get that key and we can then go forward and input put it into this file so what you can do now is open up visual studio code and what you want to do is open up folder click on this button and then find where you had cloned the actual OS copilot file so I'm going to quickly find it right over here Friday click on select opening the folder trust the authors at your own discretion go to the config.js file in this case I'm not going to be using this uh the gb24 but select whatever model that you want you can even use uh the vision model but in this case I'm just going to be inputting my key and setting it to gpt3 so I'm going to go over set this properly and I'll be right back all right so I have it set to GPT 3.5 turbo and once that is done you can just simply just click save and you can close this now once that is done you can then go back to your command prompt and you can go back to the GitHub repo this is where we can now go forward and execute your task so what you want to do is copy this command over here and they basically have stated that run the following command to start Friday replace the query which is this tab over here with your task as needed by default the task is move the ne text file containing the word agent from the folder name document to the path working directory agent now if the task requires using related files you can use this command over here so if you want to get some sort of file out of or execute some sort of task out of a file you can set this command over here now there's a couple of things I want to mention before we move forward into the analytics of this project I want to talk about the user interface where you can enhance this by starting up your own intuitive front end using Friday front end so if you're interested in that you can click on this and it showcase how you get started by deploying this Friday front end it will look a little bit easier and you'll have that left or that right panel on your screen to interact with OS copilot now if you go down a little bit more they also say that you can deploy your own API tools with fast API so if you're interested in that you can also do that and there's a lot more information on the GitHub repo which showcases how you can do that so if you're interested definitely take a look at this which will give you a lot more information on OS scope wallet and installing other add-ons to it now over here we have the OS copilot framework which is helping us build language agents that work directly in our operating system it consists of three main components you have the planner the configurator as well as the actor it first starts off with the planner which is a component that breaks down the user request into smaller tasks so this is something that helps plan to accomplish the tasks it understands the capabilities of each of the agents that will be used and it gathers information about the tools as well as the operating system to assist in planning this is then sent over to the configurator which is kind of inspired by the human brain memory system this is a component that configures each substack to help the agent complete it it has working declared and procedural memory this is then sent over to the actor and this is where the actor carries out the plan task and it has a first proposed action that is based on the configured substacks and then it actually executes them based on your operating system now after it actually executes it it evaluates the outcome and it provides feedback so it improves future executions based off of that semantic knowledge and this is how it basically operates now it goes into the Friday agent as well as showcasing the evaluations through experiments so if you're interested in this truly recommend you read through it because it's quite interesting and it gives you a lot more details as to how it basically functions and how they basically develop it so if you're interested I can leave this link as well as all the other links that I used in today's video in the description below but with that thought guys thank you guys so much for watching I hope you enjoyed this video and you got some sort of value out of it make sure you check out the patreon page if you want to access our private Discord amazing subscriptions and giveaways so if you're interested in that definitely take a look at this cuz it is truly worthwhile make sure you follow us on Twitter if you guys haven't already so stay up to date with the latest AI news lastly make sure you guys subscribe turn on the notification Bell like this video and check out our previous videos because there is a lot of content that you'll definitely benefit from so with that thought guys thank you guys so much for watching have an amazing day spread positivity and I'll see you guys fairly shortly peace out fell

Channel: WorldofAI

Views: 7,699

Keywords: open ai, microsoft 365, artificial intelligence, copilot, os-copilot, semantic memory, ai agents, ai agent, multion, you personal ai agents, multion ai agent, rabbit r1, llm, ai, ai agents examples, ai agents tutorial, ai agents explained, multion ai, personal ai agent, ai personal agents, personal ai assistant, ai assistant, chatgpt, ai technology, ai for productivity, ai application, digital assistant, agent, autogpt, microsoft ai, microsoft copilot, generative ai

Id: IVapbjYqCIA

Length: 10min 56sec (656 seconds)

Published: Tue Feb 13 2024