Hermes Agent Is INSANE (FREE!)

Hermes agent is one of the most powerful AI agents on the planet. It's free and open source, and it can build some incredible things. We want this voice agent. We have these images, this video generator, this voice creator, text to speech creator.

It can run autonomously in goals mode. And today we're going to be looking at some of the most popular and most common FAQs I'm going to break down for you so that I can answer all of your best questions about Hermes, how to use it and how to get the most out of it as an AI agent. So let's get straight into this. The first question we have here is, and these are all questions from the AI Profit Boarding, is can Hermes handle multiple tasks at a time?

Right. So if you're just getting started with Hermes, can it run on simultaneous projects? Can it do many different things at the same time, et cetera? So we're going to answer this one.

You can see there's actually five questions bundled into one here. So the first thing I would say is if you want it to work on multiple tasks at the same time, what you can actually do is create and use the Kanban board. Now, with the Kanban board, what that will do is it will take a task and break it down into subtasks. So if we say, for example, OK, go and build a website, it'll break that down into, OK, let's structure the SEO content.

Let's design the UI. Let's add a blog section, et cetera, as multiple agents working together. Now, you can have multiple profiles of Hermes, and then it can actually work simultaneously on different tasks as well that way. So that's two ways to do it.

Number one is you can have Hermes working on a Kanban board and it breaks down a big task into small little tasks. Number two is if you want it to work with many different Hermes agents at the same time, basically sub-agents, then you can have different profiles. Right here, if you're wondering how to create multiple different profiles for Hermes agents, you can just do Hermes profile create inside the terminal and then you would give it a name, like, for example, Koda. Now, the thing I would say is 99% of people, they do not need to do this.

So don't overcomplicate it if you don't want to. But that's just an idea and an example of how you can run multiple different agents with Hermes at the same time. Let's have a look at some other stuff. Should you create another Hermes agent on the same VPS or use a separate VPS?

That's up to you personally. I know many people that are using the same VPS. I don't think there's an issue with that. Can a single Hermes agent handle multiple tasks concurrently, or is it better to run multiple agents?

So if you want it to handle multiple tasks concurrently, what I would do is just plug them into a Kanban board like this. And then you can see everything your agents are working on. Bear in mind that you can have multiple different tasks inside the Kanban agent. So you could put, for example, OK, go off and create a website.

And then at the same time, you can say, OK, go off and create an AICO tool, for example, and you could dispatch multiple tasks at the same time. The reason that I recommend using that Kanban board as well is that you can manage multiple different tasks and see where they're all up to. And then anything that you've created, you can go back and view the history like you see with this task right here. And it will move along the Kanban board as you make progress.

So it will triage it, then it'll put it into the do list and then it'll be ready. And if it ever gets blocked or stopped, you can see the progress here. Much easier than just using the terminal. This is our agent operating system, by the way.

It's inside the Apache board. Next up, can you have multiple Telegram chats for different projects? Absolutely. So you can have multiple different projects and topics.

Topics are the way that you want to do it inside Telegram. What's the recommended setup if you want to switch between several projects whilst Hermes continues working in the background? The Kanban board is the one for that. And then for those already using Hermes heavily, how do you organize multiple ongoing projects without them interfering with each other?

So, again, the Kanban board can do that. Also, if it's long term stuff. So let's say, for example, you've got multiple projects running long term. Like, for example, you might have an SEO task and then you have another task that is, for example, building out a tool that you're working on.

How do you do that? What I'd recommend is you actually create skills for those long term workflows. And also you have memories. And I'm going to break those two things down.

So number one, for the skills, basically what you can do is you can say, OK, run this workflow and then save that as a skill. So, for example, if it's like SEO optimization, it could be like, OK, use this prompt for generating images to optimize and then save that prompt as a skill so that I don't need to repeat it and you improve it all the time. So that's number one. And then number two is you want to set up a memory system.

So, for example, we use Obsidian and every single project that I work on long term is stored as a memory. And then we can come back to things later when we need it. And also the AI agent can update and organize this for us, too, as well. And then we have everything connected inside a beautiful knowledge graph and we can come back to it later.

Plus, if you use Obsidian for multiple different projects, they're all organized and connected to each other. And if you use, for example, OpenCore or, for example, Claude later, they can both connect to Obsidian as well. Great questions. Now we've got another question, which is, is Hermes actually better than Claude Cowork for real work and why?

Though, honestly, like Claude, these are two really good tools. Claude Cowork is really designed for non-techy people. So if you're non-technical, then you can use something like Claude Cowork or Claude directly, and that will be a lot easier for most people if they've never used an AI agent before. It's literally designed for that.

It was built for like people who are non-technical and a bit of a beginner. If you're using Hermes agent, it's a little bit more technical, but it's more customizable. Also, the biggest difference between them is not really about whether you use it for work or not, like they both work, but really one of the biggest differences is that with Hermes, it is open source. So what that means is that it's a free open source project, right?

And also you can plug in your own API. So you don't have to use Claude's API on like Claude Cowork. And additionally for that, you can get it for free, right? It's a free open source project.

So those are the big differences in terms of actually getting work done. If you're non-technical, maybe you want to go with Claude Cowork. And if you love Claude, then you go with that. If you are more technical or you want to unlock the full power of an AI agent, even for example, basic stuff like generating images, generating videos, something like can't do that is not designed for that.

It's designed for like sort of knowledge work, if that makes sense, more like text based work. So those are the differences for me personally. I don't really use Claude Cowork at all. I do use Claude Code a lot, especially for building out big stuff like this agent operating system.

I like to use both and that's why we have them all combined inside one big mission control so they can work together and we can get the benefits of everything. Now, this was a really good question by Mohammed. So Mohammed was asking, can Hermes replace Screaming Frog? Now, if you're not familiar with Screaming Frog, I don't really use it personally.

It's an SEO tool, but I think that's very replaceable for any sort of tool, right? So for example, can Claude or can Hermes replace Atreus, for example? What I would say here is a way that I see it is if that tool has data that you cannot get anywhere else, then no, the tool can't replace it. And I'll give you an example of this.

If we go off to Atreus, right, Atreus actually just launched an AI agent for SEO recently, with Atreus, you actually get data and you get access to SEO stuff that you wouldn't find anywhere else, right? So for example, it can tell us like how many backlinks a website has and how much organic traffic it has, and it's graph estimates as well. So for example, if you look at the organic traffic there, it's actually nowhere near what's happening in reality. But that's a good example of something where you can get access to data in a tool that you wouldn't get inside Hermes.

However, if it's just for actually doing something, for example, scheduling a task or organizing some notes or building out a tool, then I would say Hermes or Claude or whatever can replace your tool. So it depends what you use it for. If it's database that you can't find anywhere else, then no. If it's just building something, then yes.

Here's another good question from France. So France was asking, how do you use tokens or subscriptions in your Hermes system to basically make sure that you're not paying for APIs and that sort of thing? So there's a few ways that you can use this. Number one, I would say, if you want to reduce the amount of tokens you use, have a memory system plugged in so that you're not re-explaining the same contacts every single time, right?

So we use Obsidian, we've got a training on the Infinite Context Engine inside the AI platform as well, and that allows you to basically reduce the amount of tokens you're using because you're not re-prompting and you're not spending 20% of your tokens just re-explaining the same context to all of your AI agents, which a lot of people do, right? So that is number one. Number two, you can actually get free APIs. Let me show you an example.

So if you're using Hermes, Hermes, Hermes, not Hermes. So if you're using Hermes, you can type Hermes model into terminal, and then you can switch over to News Portal. And News Portal gives you a free API. So right now, for example, you can get a free API.

It's called Step 3.7 Flash. Step 3.7 Flash, it's free for 30 days. And so if you want to minimize, if APIs are your worry, then you can minimize your costs with something like Step 3.7 Flash, because this is a free API you can use, right? Obviously, do you get the same power as the best API in the world?

No, but is it sufficient to get stuff done? Absolutely. So that's another option. So that's option number two.

The final option that I would say is that you can use OAuth login. And I'll give you a couple of examples of this. So in fact, I'll give you three. So number one, we use Grok and we've used Grok in the past as an OAuth inside Hermes, because I'm already subscribed to Twitter, I can log into Hermes using OAuth, which means I don't need to use API, which means there's no extra costs because I'm already subscribed to Twitter.

Another example of this would be, for example, Minimax, right? Minimax M3 just dropped recently. They've got a coding plan and you can log in with OAuth using the Minimax M3 coding plan into Hermes. And so what that means essentially is we can use Minimax with Hermes and then we can generate images, we can generate videos, we can generate voice notes, and we can actually talk to it in a real voice conversation like you can see right here.

So you see how you can get a coding plan like that, and then it unlocks a whole new bunch of capabilities that you wouldn't normally have inside Hermes without having to pay for an API. And it's the same with chat GPT as well. Chat GPT, you can log in with OAuth inside Hermes and there you go, right? So let me show you how to do that.

For example, if we were going to set that up, let's say, for example, we want to set up Hermes with Minimax, we would type Hermes model into the terminal. And then we would look for Minimax, which we've got set up here, and then you can just select OAuth, right? OAuth is a better way of avoiding API costs and this way you unlock more power without having to worry about APIs and that sort of thing. So that's a really cool way that you can use it as well.

And then also you can just train Hermes to be more efficient in the way that it does things. The longer you use it, the more efficient it becomes. So the more efficient it becomes, the less tokens it will burn through as well. So that's basically it.

I don't think that I'll say is they actually release and they are always releasing new updates that try to reduce the amount of tokens Hermes uses. So you can see an example that just came out recently, and this is basically a really interesting way that they're making Hermes efficient as well. So that's another powerful way to just reduce the amount of tokens used. So thanks so much for watching.

If you want to get this full setup for agent operating systems with Hermes agent, you can get that inside the Airprofit boardroom. If you want me to answer your questions like we've done today for everyone, then you can join the Airprofit boardroom. Just post your questions inside the community. And we create video tutorials like this every day.

If you want to get more training on Hermes agent, you can get my new daily trainings. For example, like we've got a full video tutorial on step 3.7 flash with a full guide. We have a full guide to Hermes and Obsidian, how to use Minimax M3 with Hermes agent and our full agent operating system with a zip file that you can install along with a video tutorial. Plus we have four weekly coaching calls.

Where you can ask questions about your setup, ask questions in real time, meet other cool people doing stuff like this. And also inside the map, you can actually connect with people in your local city who are using AI agents and AI automations like Hermes today. So feel free to get that link in the comments description or go to the airprofitboardroom.com. Thanks for watching.

Cheers. Bye bye.