Suna: FULLY FREE Manus Alternative! Generalist AI Agent! (Opensource)

Looks like we have a new era of general AI agents like Manis, Genpark AI, as well as Deep Agent. There's so many out there, but the only issue is none of them are open source or freely available. However, I have a new general AI agent similar to the ones that we have covered on this channel. But this one is fully open source and available to anyone without any sort of limits. Allow me to introduce Suna. Suna is a fully open- source AI assistant that helps you accomplish real world tasks with ease through natural conversation. Suna becomes your digital companion for research, data analysis, and everyday challenges. Combining powerful capabilities with an intuitive interface that understands what you need and it delivers the results for you. Now, this isn't just a regular chatbot or many of the other deep agents that we've seen. It's fully engineered to be an AI agent and it's built from the ground up with four key components. We have the backend API which uses a Python fast API service that powers the rest endpoints and through this they are able to manage the thread management as well as large language model integrations with different providers like OpenAI, Enthropic or even using something like an open-source model through O Lama. You have the front end which is using Next. js JS and a React app and it's going to obviously give you the responsiveness that you would see with Manis or many of the other agents. You have an agent docker which is where the agent runs in its own isolated environment with Docker and it's equipped with browser automation code interpreter file system access tool integration and so much more. But overall it is super easy to deploy thanks to this. And lastly, we have the superbase uh database. And this is where it handles everything from authentication, user management as well as conversation history, file storage, and agent state analytics. Just take a look at this agent live in action. It is working on this task where it was sent in and it stated that based on my website sununa. so, So generate an SEO report analysis, find the top ranking pages by keyword clusters and identify topics I'm missing. So this was a task that sent in and right away you can see that the web crawling agent is now scrolling through Suna AI and it's working on finding the top ranking pages by keyword clusters and by the end of this you'll see that it works on outputting the MD the cluster report as well as the summary. So, this is actually how you can get responses and tasks done with this general open-source AI agent. Before we get started, I just want to mention that you should definitely go ahead and subscribe to the world of AI newsletter. I'm constantly posting different newsletters on a weekly basis. So, this is where you can easily get up-to-date knowledge about what is happening in the AI space. So, definitely go ahead and subscribe as this is completely for free. Now, since this is a general agent, there's so much more to it where it can do a lot of things like doing competitive analysis, coding out full stack applications, and so much more. And this is something that is 100% free, which means that you can extend on this further, and you can install this locally without having to pay for anything cuz you can connect it with a free largeish model. And the only thing that you would need to have fulfilled is the prerequisites as well as paying for the API keys for different providers if you're using OpenAI or Enthropic. But say if you want to access and try this out completely for free, you can actually do so on the cloud where they give you 10 minutes a month to access it, which is definitely not enough, but we can take a look at it live on the cloud. And this is where I'm going to have it uh create a dashboard that gives me an analysis of my YouTube channel. So you can see that it takes my prompt and it goes ahead and works on creating a research plan and development uh plan for the AI agents to execute this task. So let's see what it's capable of doing. So right away we can see that it is creating the to-do list of what it needs to do. So the data collection, visiting the YouTube channel, the data analysis, the dashboard development where it's going to create HTML structure for the dashboard, developing the CSS styling, and it's going to also test it out in the last part where it's going to debug it and see if there's anything that's necessary to fix it up. So overall, we can see right now it has now deployed the browser window, which is where it's going to execute the task of scraping my uh YouTube channel. So you can see over here if you actually read through it, it has found all the different videos. It has found how many subscribers I have and it focuses on cutting edge advancements in artificial intelligence. So that is definitely correct. So let's see what it ends up actually creating for us.

And there we go. We now have this channel analytics dashboard and it is definitely amazing. You can see that it's a responsive dashboard that gives you good insights on the channel itself. And you can see that it did a great job in creating the graphs, the top performing videos, views by content category. You can see that it is on AI agents, AI coding tech and so much more view uh the video duration with views as well as the video performance over time and it also highlights the top for performing videos. So overall it did a pretty good job in displaying the contents and it did it pretty quickly. So I guess that's the great added benefit. And now if you want the code you can easily go ahead and export it. But that's essentially how you can use this general agent to do almost any task for you. Now say if you're interested in installing this locally, you can easily run it with the following requirements. You need a superbase project. You need Fredus as well as Daytona Python 311 for the API backend as well as the API key for your LM provider. You can also connect an O Lama downloaded model. So that way you can use this whole process completely for free. But go ahead and set up your prerequisites from Superbase all the way to Reddus. And then you can configure this by cloning the repository with this command. Going into your command prompt, pasting it in and cloning it. Head over to the Suna folder by typing in cdsa. And then you can configure the backend environments by creating the env file. And then you can set up all the different requirements and credentials by filling things like the Tavly key, your API key for whatever model you're using, Daytona, Reddus, as well as Superbase. And then you can set up the subbase or subbase database, configure the front-end environments, and then you can lastly install the dependencies. Then you just need to simply go ahead in your terminal and run the npm rundev command. And then you can go ahead and start it up and start accessing suna in your local computer. And that's easy as that. This is practically the easiest way for you to use an open-source alternative to Manis. It is pretty functional and great in my opinion. So, I highly recommend that you take a look at it. It gets you the same sort of results as you would with many of the other deep agents that are out there. So, highly recommend that you take a look at it. I'll leave all these links in the description below. Make sure you go ahead and follow me on the newsletter, join our Discord, Patreon, as well as follow me on the YouTube channel. Make sure you subscribe, turn on notification bell, like this video, and please take a look at our previous videos cuz there is a lot of content that you will truly benefit from. But with that thought, guys, have an amazing day, spread positivity, and I'll see you guys fairly shortly. He suffers.