Live Build: Testing Quifos 9B, Ornith 1.0, and Hermes Mixture of Agents

What's popping? We are ready to go. Just bear with me a sec. Who do we got here? We got Randall. Welcome Randall. All right, let's get this party started. We're consulting the Oracle. We'll get the latest headlines in a second. See what's going on in AR today. Let's have a look what we're going to work on. So, we're going to set up a new local model. It's called Quifos 9B. So, we're going to test this out.

See how it goes. Uh, I heard good things. We're going to work some more on ornith. Run some tests with Hermes mixture of experts. Check out DSpark that just dropped from Deep Seek. and also check out a new app called Open SEO which looks pretty cool. And then we'll be doing some Q&As's from the Apuffer boarding community as well and building this into the agent OS as normal. So, let's get straight into this. If you got

any questions as we're building, feel free to ask. Can you set this up in the local engine? Let's test it as well. Make sure it doesn't slow anything out. Evil Eddie, thank you very much, sir. I appreciate that. Let's go. Here we go. Playing some banging beats from the uh Hermes music agent. Hermes loves a good rave. Heat. Heat. and then what I want you to do is put those on Goldie bench. So run your normal

test but use Hermes mixture of experts. Tell me which models you're going to use for this. And then also make sure you save inside the workspace as well. Nice. create a guide on it of course as well. You set this up inside the SEO section. Have a separate tab for open SEO. Just grabbing the API key. down. Heat. Chris says, "Is this a club. It is. It's the AR profit boardroom club. That's what we have going

on here. 100% agree. Create a guide on it as well. We're getting the classic service was busy. This is crazy. This has been happening all week with Claude. Let's see what we get in a sec. We'll keep going. Sometimes you have to switch a model or whatever as well. Can you set up a new section inside Goldie Bench specifically only for local models? So for example, right now we're testing Quiffos. Um we're also testing ONIF and

you've done some tests with North Mini code as well, right? anything you've tested, put inside there. And then also run or nif in the background and start running through the benchmarks. Yeah, know it happens all the time. Honestly, if you build a lot with Claude, you'll see that pretty much every single day. Down. Hey. Hey. Yeah, if you use it a lot, you'll see it for sure. Um, I mean, there's so many, right? So, you can

use the for free models for Hermes. Let's list them out here. last last time I checked you got news portal and then with news portal you got step 3.7 flash neatron 3 ultra and they add new free models all the time now on open router you get a bunch of free APIs you can use too and all you need to do is just type in free to see the this once. So if we go to open

router here, you can type in free like this and then you'll see loads of different APIs you can plug in like north mini code, Hermes, Nvidia, loads like that. Okay, so that's that's some options and then you got local as well, right? So you could use LM Studio or you can use OAMA. So for example, if you check out North Code, Gemma 4, these are free options that are designed for a Gentic models. And if you

have a really good setup, you can also use locally Quen 3.6, but it depends on your setup. What else we got here? Yes, you got news portal. Couple of free models there. Opera, loads of free models. Local, loads of free models. And then if you already have, for example, Miniax coding plan or Kimk 2.7 coding plan or Twitter subscriptions, then you can log in with Oorth into Hermes as well. So that's another option. So those are

all your options really for free models on Hermes. I might create a guide on that actually. I think a lot of people would like to know that. Happy tail. Yeah, let's create a free guide on that. Create a guide on the free models you can use with Hermes. You could showcase what you've built with it and also talk about these specific ways you can get access to each of these. So, how to set them up, etc.

You've created guides on all of them recently, so it's easy for you to just collect that and collect it into a nice little guide. Make sure the guide is really comprehensive because you've got so much information on this, you should actually use it. You also got all the frameworks and the diagrams and everything else too. There's some on GitHub. I think no, it's just super easy, right? You got whisper flow for talking, you know. Uh, so

just to recap the question for everyone watching, Samuel says, "How are you talking and simultaneously doing so much?" So, let's have a good setup where I've got a a walking treadmill and then we got whisper flow where I can quickly talk with the AI and type out everything that I need to. And then, I mean, the AI does most of the work, right? So, it's one of those where you're good to go. Open claw versus Hermes.

Um, I mean they're both good options, but the main thing I would say is like if you're looking at open claw, it's not as smooth. And if you look at the way things are going, there's more functionality and better options coming out for Hermes and it seems to always be smooth on every update. Whereas when I used to use open claw a lot and I I really don't use it that much anymore. When I used to

use it a lot, it was very unreliable. It would break a lot. It would take a long time to fix. It just got to the point where I was like, "Okay, just going to use Hermes instead." is way more reliable. Um, I think they're both good options though. And bear in mind, like Hermes probably wouldn't be around if open claw didn't exist. So, they interlink as well. Wow, look at this. This is pretty cool. Thanks very

much. Appreciate that, Jerry. Heat. Heat. I can a sec. Heat. Hey Heat. Heat. Heat. that heat. Heat. Heat. Yeah, most of the time I go with uh with cloud-based models honestly cuz like from what I've seen I just I can't get the same outputs with local you know it's it's not frontier level. I do have systems built into the agent OS where we can use local models. Probably the best one I've tested out recently was using

Ornif. Or seems to be the best one. Uh but honestly, like a lot of local models are just kind of like a year behind. So that's the way that I see it. But at the same time, there's a lot of cool people inside the profit boardroom. Let me show you this. We're creating like local avatar videos and stuff like that with AI. And it's absolutely unbelievable. You can see an example here. Daniel was sharing his setup

using local. this is fully fully local and he explains the whole system, what he's using, how he's doing it, etc. And that was pretty mind-blowing. Like even I've not seen that. So yeah, there's some interesting ways to do it for sure. Obsidian vault. Well, that's our obsidian vault right here. So, this is what it refers to. So, it's our memory system. Second brain. So, I wouldn't run them all at once. You could use North Mini in

the background, I guess. Just don't let it slow anything down. Bear in mind, we're using qu at the same time. Bear in mind, we're using Quos at the same time. So, we don't want anything to slow down. If that means you need to wait for Grey Force, no problem. 3. Hello. Hello. probably go with Gemma 4 for that. So, a few things. Number one, I notic you keep misaligning some of the quotes. So, for example, you

see the attached. We talked about this yesterday and I told you to update your skill so it's always centered, but it looks like you haven't updated your skill. So, make sure you fix that. Also notice that you should be making this guide more visual and beautiful, right? But if you actually look at the page, it doesn't look that nice. And also, you should be having a chaty hero image inside there. you've already got the API inside

your skill. So, just update your skills so you don't forget that because you should be doing that every single time. You improve the design of the page just to make it look more interesting. Heat. Heat. anyway to get what the agent OS nice it's just like something we give for free to a profit the boardroom members just to help them really and to show them some cool stuff. So, where's the actual test for this? where you

shouldn't be using the API. I would recommend that instead you use local only. Right? If we're doing benchmarks for local, you should only be using local, not API for this. Let's see what we got here. How can I create my own agent OS? You got two options really. Number one, you just describe exactly what you want inside Claude or Hermes. Then you can start building out from there. Or number two, you can get our setup from

the AR profit boardroom. So if you look at this agent operating system that we have, we've already built it out for you and then we just give it to members inside the AR profit boardroom with all the functionality. So if you want my setup inside the air profit board community link in the comments description or go to the air profitboard.com and then just go to the classroom and go to new daily updates over here and you

can grab it's got video tutorial last update date and then also all the resources inside there. But if you want to just start building out just just start coding away inside Claude and get it to build out for you. asked Obsidian to be installed with Claude and use it, but I'm not sure if I'm using it correctly. Whoa. Just check it, right? Just check. Is it being used correctly? Is it being updated? When was it last

updated? Test Claude, ask it, right? Are you using the memory or not? Right? And from there, you can easily tell, okay, is it using obsidian or not? Is it organizing it or not? If your graph doesn't look not um updated like this or if you see like the memories are old and it's not adding the new stuff, then you know it's not working. Just go back to Claude and describe the problem. So says, "Can you be

my mentor?" So you can get that inside the AR profit boardroom, right? You can direct message me anytime. I answer questions inside the community every single day with video tutorials, the access to the whole community. So if you need stuff like that, check out the airproof for boardroom community link in the comments in description or go to the airboard.com. Is the agent OS preconfigured? Yeah. Yeah, it's ready to go. So basically what you have here is

a full setup and then you just install the zip file with your agents and then it's all ready to go. So for example, if you look at, you know, we've got a lot of members using this. Like Hockey said, "I've installed the agent OS, explored it. Your system is amazing." Rick said, "For example, I joined the school 30 minutes ago and already built the OS." You got, for example, Pobing who built out his system with paperclipip

using ours. So uh and Amateo as well. He said, "I've got my agent OS vi profit boardroom. It's amazing and just really learning everything I can do with it. So yeah, it's all ready to go and then you it's designed to be as easy to install as possible. What 191 pages of testimonials from people absolutely loving it. Thank you. No, Heat. Heat. Heat. Heat. N. Heat. Heat. N. Heat. Heat. Heat. This is a great question. Any

suggestions on lead generation? So, this is something that we get asked like all the time to the point where I was like, "Okay, here's what we're going to do." So, I actually built out a lead generation tool inside Hermes. It kind of feels like a SAS software. You can type in, you got a dashboard of like all the leads you've reached out to. If you want to generate leads, you can type exactly what type of leads

you want to generate over here. So for example, it could be like SEO agencies that are interested in link building. You type that in, you click find and enrich and then it generates a list of leads like so with the contact details and then it gets enriched automatically as well. So you can see inside the notes section here it's got details on every single lead that we've generated inside the system. Now you can delete any you

can add any to a campaign like slow so and then once you've done that go to campaigns you can start doing the outreach and also there's a a send box in the inbox for using this as well right so that means basically you can automate the lead generation you can manage everything with Hermes you can manage the campaign and what you actually send and then you can manage your inbox and send inbox and then if you

want to change any settings you can add like a hunter API key fire API key etc. inside this section as well. So, you get everything you need to win with lead generation. Um, and that is a a pretty amazing system that you can use with Hermes agent directly. Next question. I saw chat GPT5 in uh 5.6 in there, but that has not been released. So, is that just branding? Where did you see chat GPT 5.6? We

we've not got it inside our system. What we do have though is we have a we have a guide on it that we talked through yesterday on GPT 5.6 which is here but we run through like you know is gated and you can't get access to it yet blah blah blah and we already got a full guide on that but yeah it's not available for anyone. Uh I think there's only like 20 people that have access

to it. 20 companies so not everyone has access to GT 5.6. Yeah. There you go. All right, let's see what we're up to with these benchmarks. talk about some of the workflows and show examples. Be clear that the data might not be as good as other tools. I would just test it yourself and see what you think. Even the best tools don't have the best data to be honest with you. There you go. Yeah, I think

so. for now. Like that's a very basic CRM. And then if you're like, "Right, well, I need to build more functionality into it or I need to tweak it how you want, you can always use this system and then customize it more." So you can say to Claude or whatever agent use to set it up, okay, listen, add that or change this or tweak this or whatever you want to customize inside any part of the agent

OS, you can. That's the beautiful thing about it, right? Like it's it's yours to mold and to change and to switch around however you want it. Just a heads up as well. Can you make sure that you include a workspace where we can see everything we've created with this tool? Thanks very much. Yeah, I tried to make it as nice as I possibly can, but as simple as we can as well. Bear in mind as well,

like anything you see inside here, you can switch off. So, if you go to the settings, you can hide anything that you want to. So, for example, like we hid the the local agent camb just cuz I don't use it that much. So, you don't have to use all of it. You can just use the bits you like. You make sure you do all 42 tests. Just reconfirm it as well. Can you show with proof that

we actually used Mixer of X agents inside the setup and also inside the guide includes some comparisons to Claude 4.8 Okay. So then people can get feel for okay how does it perform versus other models? How does it perform versus fusion for example? But I would do that for the more bigger stuff. So for example like the dragon realm game or the cry game. Well, I mean, it's probably not relevant cuz I test everything, right? But

if you want to use free models, there's so many different free models or local models or free APIs you can use that you don't need to worry about that or you can use your existing CLI as well. So it's not really relevant what I use because I'm literally testing everything and benchmarking everything. So um but even then like the the amount that we use is not that bad. Not that bad at all. I mean if we

have a look for example inside Claude building this whole system we're nowhere near our uh we we came nowhere near to hitting the limit last week. So it's all good. I noticed when you've done this, you've not added any animated workflows and you should be right. Also, I'd recommend like putting the tool right at the top just so you can showcase it. Update your skills so you never forget to do animated workflows and diagrams cuz that's

a big part of explaining how this works. For each example, you should embed the tool and showcase like how it can be used. That way, it's really visual cuz you don't want to just tell people. You want to show people, right? Update your skills so you never forget this. Thank you so much. I appreciate that. I mean, bear in mind like the the avatar helps as well, right? Heat. Hey, heat. Hey, heat. I notic that the

CTA is so basic. Like, it doesn't even have a proper CTA walking through everything that you get with the Agent OS and it's not really expensive. So, just update your skill so you never forget to do this cuz right now it's super brief and it should be a lot longer. You should have two CTAs throughout the page. Update your skill for guide so you never forget this too. As you complete them, can you add them to

the guide, please? Just do one by one horizontally so that we can see each one and preview it, but it's big enough to see probably computer use. So sometimes like the AI will use computer use and that's why you'll see two cursors. I'm going with Hermes. Hermes and Claude are right up there, you know. Uh, so if I have a choice between Hermes or Codeex, I would go with Hermes all day. And then if I had

a choice between Claude and Hermes, I would go with Claude All Day, but you can use them all together. So for example, you can plug in Codeex directly into Hermes, and that way you can get the power of both of them. The other option you have is you can use codeex inside the agent OS, and you've got Hermes there as well. So you can use all of them together if you want. Down. I can't see anything

inside the guide showcasing what we've actually created. You need to put your tests inside there as a showcase just like I asked before a second ago. Rufflow is free and open source. Yes. You can use it with CLI as well. back in a sec. Heat. Heat. N. Heat. Heat. Heat. Heat. Hello. Hello. Thanks for joining. Hitting limits like crazy with Claude. Interesting. I've not had that issue at all. It depends what you're doing, I guess. But

yeah, so far for me, it's been fine. Welcome here, Jay. Good to see you here again, mate. We're going to run through some a proper boarding Q&A in a sec. Let's have a cheeky little stretch. Wow. All right, let's see what we got here. Oh, this is great. Wow. This is uh using Hermes mixture of experts, mixture of agents, sorry. It's pretty wild. Let's have a look at that one. Can you have it more like the

workspace with Hermes workspace instead of the showcase that you've got on the builds page of mixture? That way it's easier to preview and see everything. No, I can't. You can't use Hermes CLI with Claude at all. Yeah, I saw that too. Claude have increased limits recently. on Goldie Bench. Can you make sure that you have the same layout on Hermes mixture of agents page as you do on every other page? And then also set up the

compare pages and everything else. And then also update the leaderboards too based on what you've done so far. Let's see what else we got here. So, we're just testing out Hermes mixture of agents here. Do you have the guide as well? Where you up to on the CTA? I notic that you put the local models like orif on the leaderboards, but you probably shouldn't. What I'd recommend instead is that you put those inside a separate local

leaderboard because it's not fair to compare local models versus Frontier models. So you probably want like a local benchmarking system instead. Nice to catch a decent time. Could see here too. Well, it depends what you're building. Only you know that, right? Only you know how much you use. But for me personally, like Claude is amazing. again like if I had a choice between Claude and Hermes, I'm going with Claude every day. What's the difference between Hermes

workspace and desktop as per your use cases? I would recommend that I I've done a video on this, but I would recommend that you go with the agent OS instead out of all of those because you'll get a lot more customization. It'll be a lot more powerful and there's just so much more you can do with it that you can't with anything else. So the other problem with for example if you're using Hermes desktop or Hermes

uh workspace you can't use your other agents right it's only for Hermes. So for example if you want to have teams of agents like claude and Hermes working together you can't do that with Hermes workspace or Hermes desktop. If you want to plug in uh memory galaxy or customize it like this you can't do that with with Hermes desktop or Hermes workspace really. Right. So the main point here is like you really want a system where

you have everything built in and then that way you're not relying on one agent and you're not relying on one model and you can switch between them and I think that's the smartest way to use AI in 2026. All right, let's see what we're up to with quefos. Think that's still running or if is still running. Open SEO running. Per M of agents were running. Okay, I I'll tell you what we'll do. We'll do a a

little community Q&A. So today we're going to be answering the latest questions on our agent operating systems. This is a powerful system that plugs in paperclip AI agent group chats uh idea to implementation pipelines for apps, websites, tools or landing pages you want to build in one single click. We've also built in some new functionality. So, for example, if you want to generate leads now, you've got the email and lead generation section with Hermes. Over here,

you have Hermes Oracle where you can pull in the latest news. You've also got Hermes Jarvis that you can use to basically voice control your AI agents. You got studio for generating everything here. And a couple of new features that we've added. We might have to start that one again. It's rebuilding the workspace. I can see it working on that right now. We'll come back to this in a sec. Wow, this is awesome. What on earth?

The Hermes mixture of agents is is pretty wild. This might be a new number one on the benchmarks. It's looking awesome. Yeah, I did. Yeah. Yeah. And there's a tutorial on that in the air for boredom as well. Today we are going to be running through the agent operating system that I've built, some of the top questions I get on it. I'm going to help you as much as I can on this sort of stuff. And

I've built some absolutely amazing stuff into this recently. I want to talk you through exactly how we built all these different basically I I mean these are are some of the best responses I've ever seen and this is using a new feature we've just added called mixture of agents from Hermes agent which is pretty wild. We also got the Hermes Oracle inside there. So we can pull in the latest trending news and grab it straight away.

We've also, for example, got a lead generation system in here for managing our emails, doing outreach campaigns, and generating leads in whatever niche we need to. And then we've got things like the memory galaxy that gives all of our agents contact straight away. We also have a a music agent over here. We have a video agent that can generate amazing videos. We've also plugged in uh open montage, which can generate like basically cinematic AI videos as

you can see. Uh, and these are all powerful systems that we've plugged into the agent operating system to automate and build anything that we want. And so what I'm going to do today is just answer some of the top FAQs from the air profitable community because I know if everyone else inside the community is asking questions about these agent operating systems, you probably have the same questions, too. So, I'm going to get straight into this and

talk you through everything that we're using. So, quick one from Louis. Lewis was asking about like custom agent builds. Do we do that inside the air boardroom? We actually do. So, we have an option for agent requests and this just a thread you can post your requests on. There's no limit to that. So you can just post whenever you want and as long as we don't already have the agent built then we'll build it for you.

If we do already have it then we'll point it directly to you. Next question from Danish. So Danish was saying you know with the agent OS pipeline the SEO pack which is down here. This basically allows you to like generate and automate SEO with all the systems we built and we've just added open SEO inside there as well which is pretty cool. So that can do your keyword research for you. It can uh rank uh it

can do rank tracking for you. All sorts of cool stuff. Right now, if you're wondering, okay, do you need Hermes to use this? Well, basically, you can use anything you want. Whatever API or CLI you prefer to use in this, you can. So, you just ask the agent when you set it up to use that particular API or CLI. So, for example, if I want to use Claude CLI instead of Hermes for the SEO section, I

would just ask the agent that built it to set that up for me instead. So, it's really flexible depending on what you need. The cool thing about this is like we can grab keywords from our Google search console. We've got a free SEO tool for managing everything here. We can generate content and publish it directly to our websites. And then we can actually get SEO traffic using the system. And it's working really well. Let me show

you a couple of examples. So these are some of the websites that we've set up using this system. Here's one. Here's another one. And here's one more. Right? And you can see the trajectory of all of them. Like they've gone from zero to hundreds of clicks per day. So it's a really powerful system that's working for us very nice. Now another question from Danish is how can you feed Obsidian memory to Hermes? So because Hermes is

set up locally and so is Obsidian. You can just give the local directory of Obsidian with the documentation about Obsidian from their website directly to Hermes and then that will plug it automatically into Hermes for you. And then you can say, okay, every time I use you, please update the Obsidian memory or once a day consolidate all of your memories and plug them into Obsidian as a new daily log. So just give it the local file

directory. And then for the section where you say, you know, do I have to specifically say whilst building the article writing function to use obsidian memory? Yeah, I would if you can. So I would just say like listen every time you create content personalize it to me and make it unique to me with my unique goals, my unique frameworks, my unique case studies using the data inside the obsidian memory. That way we can add information gain

inside our content so that it's the best it can possibly be and unique to us. Great questions from Danish there. Jay was asking about the agent camban and the local agent um the local engine section. So this is something that we've built in where you could use local systems. So for example like this morning we set up quifos which is a new model that you can run locally with your AI agents. You can see examples of

it working right here. So the way that you can fix this is you can just give the details of whatever local model you prefer to use to the agent that set up agent OS for you and just say hey use this model instead or use that model or every time I install a new model change the default model to that new version. And this way it's easy to swap and interchange between whichever local models you prefer

to use. Another question we got asked was about free claw code and omni route. Now this is not something I've tried personally. I've just taken a look at the example get up here. I mean you can use whatever you want. I think there's many different ways to use whatever model you want inside claw code. For me personally, we just use the free clawed code GitHub, but you could swap and replace it with whatever you prefer. So

if you find like a a GitHub setup that you're like right okay I prefer that setup for free code then you can plug it in here. So personally I've not used it but feel free to test it and if you prefer it you can swap it in and out. The thing I would say here is there's many ways up the mountain. So whichever way you prefer feel free to customize it as much as you need to.

We've got a question here from Amosu about OMI. Now, OMI is what we use to basically update our Obsidian database. So, what this does essentially is it takes notes on me all day. It's like listening to my conversations. It's watching what we're working on, etc. And then we have a bunch of memories that it creates daily. So for example, if we have a look at memories here, it's literally looked at what I've worked on recently. So

we worked on working on Goldie Bench, an AI modeled leaderboard. And then it's taking notes on that, right? So it's constantly giving Obsidian more context. And then if we want to export that to Obsidian, we can go to more here and we can easily obsid um we can easily sync this to Obsidian like so. Pretty simple and easy. And then once we've done that, that's going to be inside our memory. As you can see here, we've

got 3,000 memories. And this is all plugged into our HNOS. And the beautiful thing about that is that it's basically an automated system to make sure that your your memory and your second brain is automatically updated and organized every single day. Now, if you want like a a fully free version of that, that's an alternative. We actually got a tutorial on this right here, right? using this GitHub and this is a free local version of that.

So if you want a different way of doing this, I would recommend that But yeah, you can set this up locally and that's a good alternative and pretty easy to set up as well. But it basically does the same thing. I just don't use it as much because it's not quite as smooth. So, we got a post from a le leg legendary member inside the group, Jay, who's basically saying about how like he's using Claude projects

for organizing everything that he does. I think this is quite a nice way to do it. So, if you if you're not familiar with Claude projects, there's a couple of ways you can do this. So number one, you can go to Claude inside here and then there's a project section. And if I'm sharing this with a team, this is actually how I prefer to do it. So I have like a team setup that my team members

can be added to and then they can get access to all of these different automation projects that we have right here. And also all has context on me. So for example, if we go to this section here for the email writer, it's got the previous conversations. It's got custom memory, custom instructions, and then we can add examples of what type of emails we like so that my team can easily automate and create them. But yeah, you

can also use it directly inside Claude code as well. So if you go to co-work, there's a project section here. I don't tend to use that much. I just tend to use the claw code section. And then when it comes to building stuff, I would just use uh the code section here. Also, his tip is that at the end of every session, Claude gives him a daily build report of what was built, what was changed, and

what's next. And then it goes straight into the project folder, which is cool. has just joined as well and they were saying basically they started with openclaw back in January and then moved over to Hermes agent at pi and never looked back. This is exactly the same journey I've been on this year where I was using openclaw. It kept breaking. I was spending more time fixing it. And so if anyone's watching this and you're thinking, okay,

should you use Hermes or should you use OpenClaw? For me personally, I found Hermes a lot smoother. Every update seems to work and the functionality they're adding now is just absolutely amazing. I'll give you an example. So what they just launched yesterday was something called Hermes Mixture of agents. And we've tested this out and it's absolutely amazing. So what you can have is like multiple models working together. As you can see right here, we've got a

panel working through open router with Hermes. Then we can ask the panel to do anything. And you can see some examples of what it created. Like it created a full OS over here. It created this um amazing like running game. So you can see right here like pretty awesome stuff that it's generated. And this was super easy to do as well. So we've just added that section inside our agent OS where you can basically use mixture

of agents build anything that you want and then it stores it inside the workspace here and we've got a full guide on it on the on Goldie bench. So you'll find a mixture of agents over here. This is an interesting one. So Neil is trying to use Olama with GLM 5.2 and claude code. Let me just check it out now and we'll test it ourselves. See if it's working. So we can run it as long as

we got lama running. We can just run this terminal command with Lama launch claude model GLM 5.2 cloud. Let's see if that works. Yes, that that's working fine for us. So, it might just be your setup. I would get Claude to actually go in and fix it for you if you can't get it to work. Also, just make sure you have OAMA running in the background. And also make sure that you have updated to the latest

version of Olama as well. And then if you're still struggling after that, I would recommend that you actually get Claude or Hermes to go in and fix that for you. So that's basically it on the agent OS and the community questions who've answered everything today. If you have any questions you want to ask me, just join the AI profit boardroom link in the comments description or go to the aiprofitboard.com and then I actually create video tutorials

daily answering all the questions inside the community so that you can ask me questions and get help and support whenever you need to. And also you get access to the whole community with 3,800 people inside there. So you can get help and support 24/7 pretty much. Uh if you're wondering, okay, what do you get inside the agentware? So you have uh mission control, you can plug in Hermes, Claude, whatever CLIs you prefer to use. If there's

something you don't use, you can always hide it over here with the settings. And then obviously you can get all these different systems that we've built in based on members feedback. So based on what people want to say, see, we build them in. So for example, like people are asking like how do I, you know, ship things quickly? I I have shiny object syndrome, whatever. So we created this system where you can drop an idea, create

it and then you can literally go from like idea to build with your AI agents automatically and you can see everything that you've created and store it in one place which is awesome. You can also see for example we've got Claude, we've got Ultra Code built in there as well. So you can use Ultra Code whenever you need to if you prefer to have swarms of agents working together. We got Hermes agent where you can chat

with it, you can talk with it, you get a voice activated version of it inside this section and you can see the full conversation plus you got all of the stuff that you built with this as well which is pretty cool. Over here we have Hermes Oracle which pulls in the latest news and this automatically refreshes every day every 24 hours so that you get the latest news pulled in. We have a studio where you can

generate videos, images and voice an outreach system for lead generation and emails. um the mixture of agent system that literally just came out like 24 hours ago. We've already built it in. And then if you want to add MCPS or see what you've built, etc. That's in inside this section right here as well. And then we've got like all these different CLI. So if GLM 5.2 comes out, well, we already plugged it into the system like

ASAP. And we can see everything that we built. Uh when for example like Fusion came out, we plugged that straight into the system. So we can get like Fable 5 level intelligence. We got a local engine here which we've just tested with Quifos 9B which is pretty cool. Then we got Sakana Fugo as well which is a Japanese AI on benchmarks achieves similar levels to Fable 5 as well. Then we have all of these different automations

and workflows based on anything that you want to automate here. So whatever you want to build even for example like if you want to create films where we've got open montage here with AI animations and and the quality of this stuff is just wild like insanely good. So this is all inside the AR prof description or go to the prof.com. Inside the classroom, you can get access to the agent OS over here. And we update this

daily with video tutorials. Plus you can get the file for installing it. We have new daily guides on everything that drops is actually useful. So for example, if you're interested in mixture of agents, we've got a full setup right there. Inside the community, you can ask questions and I personally answer them every single day and so does the whole community. You can see, for example, there's 90 people online right now. So, there's always people online ready

to help you. It's a very active community. Inside the classroom, you can get access to all of my new best trainings. Inside the calendar, you can jump on weekly coaching calls, get help and support in real time, share your screen, ask questions, meet the community. And then inside the map, you can actually meet people in your local area who are building with agents or building with the agent OS and that sort of thing. You might also

say, okay, well, doesn't this take up a lot of tokens to use this sort of stuff? for example, if you want to use the agent OS. The answer is no because you can use free APIs if you're worried about that. You can also use the CLI you already have. And then additionally, you can also use free local models as well. So there's three free ways to use this for free. And then also you might be thinking,

okay, this sounds technical, it might be difficult to set up, etc. We've had so many members set this up and get awesome results that I know this works for everyone. So for example, Hockey, he installed the Agent OS. He said the system is amazing. Rick, he set up in like 30 minutes. Gil absolutely loving the weekly coaching. Hobing just set up paperclip inside his agent OS. Uh Raza said we're one of the best trainers out there

for AI. So if you want to win with AI, feel free to join us inside the air profit warning. Thanks for watching. got a question here. What is a benchmark and what do you mean by number one on the benchmarks? And so a benchmark is a way of testing a model to see if it's actually good or not. And so we rank all the models and we test them out. We see which one performs the best.

And then based on that, we take a look and we're like, okay, this one is the best or that one is the best, whatever. And so it's just a way of comparing different models based on testing and seeing what performs the best. How did you get Quen? You can you can just get the local model. It's pretty chill. So today I'm going to show you how I build my own SEO tool essentially using Open SEO and

this is a free open- source project. We've already plugged it into the agent OS inside our SEO section and this looks really cool. So what we can do here essentially is we have our own open- source tool and for example if we want to automate keyword research we can easily do that over here. So for example we type in Hermes agent we can change the country we can change how many results we get back whether we

want it to be auto related suggestions or ideas and then we hit search and this is literally like a fully customized tool that we have ready to go. So if we check out the search trends here it shows us the search volume. We've got all the different keywords that we can plug in. We can actually click on those and then save the keyword as you can see right here. So we can save that inside our system

or export it to sheets or CSV kind of feels like a dress a little bit. And then we've got the SER analysis over here, right? So what this means essentially is like we can quickly find new keywords relevant to our industry. We've got the details of the volume, the competitiveness, the score, what type of search intent actually is, which is great for SEO. We can filter all of these keywords and this is all driven with AI.

Um, we plugged in a API called data for SEO into this. And then from there, you know, you don't need to subscribe to an SEO tool. You've got this ready to go. If you have a look at the SER analysis as well, it shows us all the results for that particular topic, which is great. And then we can actually switch here between keywords, save keywords, and we've got a rank tracking option here. So, what we can

actually do is we can plug in our domain like so. set up the country, whether we want it to be like desktop or mobile or just mobile or desktop, how often it updates, and we can create our own like SEO audit tool over here whenever we need to, which is pretty cool. And then it pulls in the number of keywords we want to track. As you can see, you can save those keywords, and it will automatically

track each of the keywords for you. Now, if we go to domain overview, what we can do over here is we can check a domain. So for example, we could plug in like Julianolder.com, search by traffic or rank or volume or score. Hit search and then we can actually analyze the domain itself, which is pretty cool. So it shows us all the keywords that a competitor will be ranking for, where their pages are getting traffic, what

sort of keywords are ranking for, what position they're in, uh the amount of volume of traffic they gain, the tra the score as well overall. So that's really good for analyzing your own domain and also analyzing competitors. So this is a fully AI powered system and the cool thing about this is we can just plug it into our agent OS. So if you have an agent operating system where for example like maybe Claude or Hermes or

Antiggravity will live. Well, the great thing about that is you can go back to your open SEO section here and you can see all of your history, right? It's all saved over here. We have the domain. We can also check for backlinks too. So we can analyze a website, enable backlink data here. We have AI visibility. So we have like an AI rank tracker inside the system too. And then we can also pull in MCPs here.

So for example, we could have an MCP with Claude and add the CLI for Claude code which is pretty cool. A Claude desktop or Codeex, whatever you want this to run with as well directly. We also got a bunch of skills that we can add into Codeex and Claude for reusable SEO workflows. So for example like researching keywords, getting tracking positions, getting keyword metrics, etc. This is super powerful. Now if you're wondering where does this come

from this is the open-source setup. So you can see it's called open SEO on GitHub. It's free. Pretty powerful system. It just it's kind of I mean it if you look at this it's like okay you can do link prospecting. You do competitor analysis. You can do keyword cluster and keyword research. You've even got an SEO coach built in there and an SEO project set up as well. So there's a lot of agentic SEO skills built

into this. But the difference here is like it's not just for example like going into Claude and saying, "Hey, give me some keywords." This is a system. It's it's a tool. It feels like a SAS software almost when you're using this. And also it's got rank tracking which would take ages to set up manually if you're trying to figure out that yourself. And also the cool thing about all of this is you just plug in your

API key and then you can go from there. You can also add and manage projects. So you can have multiple different projects for different websites or clients you're working with. And then the cool thing about this is as well it can link straight to our SEO workflows already plugged into the agent operating system. So for example we have a look at Hermes agent and we see okay what are people ranking for in that particular niche and

what sort of keywords could we go for. So if we scroll down here let's see what we got. Yeah, this is a good one. Right. So, you can see this keyword right here. So, what we can do with that keyword is then we can go to our generate tab which is something I've custom built inside the agent operating system and we can take that particular topic. We can get a case study for it or we can

pick an existing topic and case study for example like this one. And then we can from there click on generate five articles and based on the keyword research we've automated with open SEO. We can then start generating the content, plug that into our systems and start ranking with it and deploying that automatically. And this is great cuz it's just like oneclick SEO, right? I don't need to log into WordPress. Super easy and simple. It's all plugged

into one system and it's all organized inside one place. Can you imagine for example if you're going over to Claude and then you're going over to HS to get the data and then you're going after search console to check your rankings and then after that you're going into WordPress itself to start posting the content like very very messy a lot a lot very energy draining a lot of like logging in logging out and clicking buttons or

you could just fully automate it with this system as you can see right here and I think that's a much more powerful way to use this whole system and then we can see the history of everything that was deployed, all the content that we've created over here. And how easy and simple is that, right? And if you're wondering, okay, like does this actually work? Let me show you an example. So, if we go to some of

the pages, some of the websites we're working on right now, like these, you can see our traffic over here, right? It's it's on a really nice trajectory since we've started using systems like this. So, I think the more you can systemize SEO, the better it's going to be. Obviously, you have to quality control this stuff, which is why we've got a 13step skill inside the agent operating system for SEO to quality control it step by step,

but SEO has never been easier and it's never been more fun. And you can easily plug in systems like this and then start ranking with them. So, that's the system that we're setting up. Also, what I want to show you here is some examples. So, for example, I could actually plug that into a website that we're building here. We can then create like a free audit tool for people who join the website. And that's that's a

pretty powerful lead generation method using the systems from Open SEO. We could, for example, use this to generate more leads, to drive more traffic, and to build a website around this, too. So, that's some examples of what we're building here. Here's another example. So, this is a full guide I've created on Open SEO and how to use it. And the way this works is like Open SEO is a free open source self-hosted SEO research tool. And

then we've wired it straight into the agent OS with its own tab, keyword research, uh rank tracking, and backlink setup as well. And then the point here is like instead of having a monthly subscription and not being able to customize your SEO tools, you can customize this as much as you want and then you just use something like the data for SEO API and you change that about as much as you want. So for example, you

could use this for keyword research. Here's an example. And then you get the keyword, the volume and score, the SER analysis. You can automate the content and you get better rankings for that. So, how could you use this? Well, for example, you could use this for finding the easy wins when it comes to SEO with the keyword research tool. You could check the SERs before you write with the automation and just quickly check, okay, what's already

ranking on the first page of Google and then how can we reverse engineer that to make sure our content is better. We can also use this to size up our competitors. So with the domain view, you can drop in a rival site, see if the keywords are ranking for, target the ones they're winning that you're not yet, and you can enter any domain into the panel to start running it. Additionally, you have you can track what

matters. So you've got the rank tracking section here, so you can track the keywords you actually want to rank for and where you're up to on that. And then additionally, you've got your AI visibility. So this setup means that you can do a brand lookup and you can see how your brand shows up across AI answers. So obviously chat Google AI overviews this is the new surface that most tools ignore but where a lot of traffic

is quietly moving and you can set that up inside the AI visibility section. And then finally we can pipe it into our agents. So you got the prompt explorer and the MCP server. So your agents can actually pull this data in automatically just by setting up a MCP. So you could for example give the MCP to Claude or to Hermes and then they can pull in that data and you can automate SEO with them directly. So,

pretty powerful stuff. And then we've got the SEO pipeline that we custom built as well. So, we can take the data from Open SEO, plug it into the SEO pipeline, deploy it across our websites. And if we go back to the websites we were looking at a second ago, these ones over here, let's check out the history section. The content that we just created a minute ago, so you can see 4 minutes ago, that is still

running in the background. And all of these articles will be written and automated whilst I'm talking to you. So you might say as well like you know doesn't this use up a lot of tokens? Well you can use free APIs for this and then for data for SEO it's it's a lot cheaper than like your normal sort of SEO tools. Also you might say well self hosting is is too technical for me but it's just like

one docker command and then it's already inside the agent OS. You might also say well free tools probably give worse data but usually it's the same data. big tools are using anyway if that makes sense. So that's the whole setup. Now if you want my entire agent OS done for you open SEO is just one tab. The real prize is the operating system. It lives inside the one I run my whole business on. So you can

join the profitab and get all of it ready to set up and run. So you get the open SEO and SEO pipeline, the local Hermes engine, agent camb for a team of SEO agents working for you, the clawed workspace with everything saved, every top L uh CLI plugged in, free local models, the AI mastermind where you can have a group chat with all your agents. You got the memory galaxy that already knows your business and all

your agents have context. Plus, if you're writing SEO content, this can pull in the context from your memory so that you get the best answers and the most unique content possible. And then also we've got token efficiency playbooks inside the AI profitable boardroom if you're worried about using up tokens and you can connect with me plus the whole community. So feel free to get that link in the comments description or just go to the aprofit.com and

inside the community you can get help and support. I personally answer the questions inside here today. There's a lot of SEOs inside there that are uh actively like posting about cool stuff that built with this. inside the community classroom you can actually see there's a full section on AI SEO and if you want to get the agent OS along with new daily updates we have that inside this section so you can get the agent OS with

the last update date a video tutorial and the C file to download and we update this daily plus you get new video tutorials based on what's actually useful and working inside the calendar you can jump a weekly coaching calls get help and support in real time ask questions meet the community inside the map you can meet people locally in your area who are using AI agents that's all available inside the AI profit boardroom link in the

comments description or just go to the aiprofit boardroom.com. Oh, Hey, hey hey hey. Heat. Heat. N. Hey, hey hey. Heat. Heat. Heat. Heat. N. Back in the sick. Hey hey hey. Hey, Hey, hey hey. Heat. Hey. Hey. Hey, hey hey. Heat. Heat. Heat. Heat. Heat. Heat. N. Enough of that. All right, let's get this party started. Today I'm going to talk uh some of the best free models that you can use directly with Hermes. I'm going

to talk about four different ways that you can use Hermes agent for free and I'm going to rate them and talk about, you know, which ones are the best and how to use them and how to get the most out of this. So, let's get straight into this. The first method that you can use is news portal. So, for example, recently Step 3.7 Flash and Nimatron both came out for free with Hermes Agent. Now, if you're

wondering how to use these with Hermes agent and news portal, what you can do here is you can go inside your terminal and then you can select Hermes model and from there you select news portal. And if you're on the free plan, you'll see a list of models as you can see right here. Right? So you can see see between all the models that are available and which ones are available, right? And quite often they bring

out free models. So for example, recently step 3.7 flash was actually free and Neatron 3 Ultra was also free. So just watch out for those. You can also compare the prices as well. So, if you want to get just a cheap model instead of a free one, you can go with that option. Now, the next option that you have is that you can use open router. Right now, if you want to change the model inside Hermes

to open router, you can just type in Hermes model again and then you switch to open router. Now, you would just plug in your API key. Once you've done that, you can select between the models as you can see. Now, also one thing to note here is that you can enter a custom model name. And if you go over to open router directly, if you type in free inside here, you can see there's a bunch of

free models. So for example, Hermes 3 405p instruct is free. North mini code is free. Llama is free as well. So you can switch between these models. Choose whichever ones you want. There's also a free models routter on open router as well. And so you can use the API from open router. Now, if you're wondering how do you get the API key if you just go to your home section here. So, click on the settings and

then home. From there, go to API keys and you can just grab an API key for free from open router and then plug that inside your terminal to start using it for free. Right? And either way, whichever model you use, there are ways to basically never have to pay for Hermes ever again. what we actually do to organize this and I'll tell you halfway through as we're getting into this. We've covered two methods already which is

new sport and open router. If I was choosing one free model by the way of open routter I'd probably go with Gemma 4 or our alpha but there's loads of different options you can try and you can test them out for yourself and see which one you prefer the most. So what we have inside our agent operating system as you can see right here is that we have separate profiles for each of the APIs that we

plug into each of these models. So the cool thing about this is for example if we're using North Mini and the free API from North Mini well we can have a separate agent profile just for North Mini we can plug in the free API into that and then if I'm like right I just want to use Hermes quickly for free I can just go straight inside here. The same for example if we go to our local

model setup so like for example or if or is another free local model that just dropped this week. It's a self-improving model and then you can use a self-improving local model with a self-improving agent which is Hermes and you can use that for free as well. So just proving number one this works and number two you can actually organize these into three profiles and then switch between the ones that you like the most. Now speaking of

local models you might be wondering okay how do you plug local models into Hermes? So there's a couple options to this method number one is you can go to Olama download O Lama. Once you've done that, you can then go into any of these models. So, for example, if we look at orif, which I was just talking about a second ago, we can then go straight inside here and we just need to run ornif. So, make

sure you have lama open and installed. Make sure you have it updated and then from there you can copy this command, go inside your terminal and just pull the model in. So, you start a new command and just run that model. Once you've done that, then you're going to go to Hermes model over here. And from there, you can scroll to Olama, which we've got over here, and we can select our options. The other option that

you have as well with this is you can use LM Studio. So, you can use LM Studio as a provider. And the cool thing about that is if you're using LM Studio, you can actually see which models are best for your local setup and it will give you a grade in terms of which models are too big and which models you can actually run on your local setup. So if we go to LM Studio, which is

free to download here, and then we click on the model section, we can scroll through and we can see which ones would actually work. So you see how it says likely too large on this particular local model. We can just keep scrolling through finding local models and find ones that are actually not too big for our setup. And that's how you can run local models. And then to set it up with Hermes, you can go inside

Hermes model and just change the model to LM Studio or OALMA. Now, if you're wondering, okay, how do you create the profiles? Like how do you have separate profiles you can come back to like this? So, we've if you go to your dashboard for Hermes agent, then go to profiles here, you can create a new profile for every agent that you want to use. Uh, another little tip here as well is you can use something called

blank slate. Now, blank slate is a way of setting up Hermes with like no existing tools. So, for example, if you're using free APIs and you want to limit the number of tokens you use because obviously free plans are quite often limited with the number of tokens, then you can use a blank slate setup with Hermes agent and that way you don't use any uh too many tokens every time you use it. So if you go

to create here, you can just call it okay free local and then just from the model section here, you can just choose the model that you want to use. So for example, you could use Lama Gemma 4 and run that for free, which is pretty cool. and then I have them organized inside my agent OS. So I can just see all the profiles I've created, flick between them and easily come back to them whenever we've created

them. So just to recap here, we've talked about news portal and the free models that occasionally go on there. We've talked about open router and the free APIs in there. Honestly, I think out of all of these, probably open router is the easiest. So if you have a choice, I'd probably go with open router first, then local, and then news portal. And then the final method is oorthth. So if you're wondering okay how do you use

o off free so if you already have like a chat GPT subscription for example you can use that directly with Hermes agent without paying anything extra. So for example, if we go inside the terminal here and we go to Hermes model, we can then select from the list and you just press down to switch between the list here and you just go to open eye openi codeex use existing credentials or you can reauthenticate with a new

oorthth but basically that means that we can select between the models so we can use GT5.5 and now that's selected as our default model with Hermes agent as well right and so that is method number four the same for example if you're already subscribed to Twitter you can use grock af and then with gro on Hermes you can generate for example images or videos or voice or whatever you want right using the the combination of grock

and the o login with Hermes agent. So that's four different ways that you can use Hermes agent for free, which is pretty cool. And that way you don't need to worry about tokens. You can also install stuff like, for example, headroom. Headroom is a way to reduce the amount of tokens you use with, for example, like Codeex. And also you can use blank slate with Hermes agent to set up a system where you don't use too

many tokens cuz you haven't installed too many tools to your individual Hermes agent. And so if you look at the old way it's like you know you have to pay for APIs you have to worry about which model you use then you use a model and you're kind of locked to one vendor and also your private data goes to someone else's cloud. Also, if you're using API and you're paying per API usage with Hermes, sometimes you,

you know, the token usage can get out of control. With this system, you can use like $0 per token on free and local models. You can loop all day if you want. You can build stuff all day, especially with local models. Your data stays private. You can swap the models based on the job. And you can reuse the plans you already have. For example, like chat cheap off. So, that's the whole system here. That's how to

use it. That's how to run Hermes agent for free. um works pretty nicely. If you want all of this, all of these free options wired into the agent operating system that we've set up. You know, you don't need to set this all up by hand. Inside the agent operating system in the ARI profit boardroom, we have Hermes and all of our agents plugged in so that you get Hermes, our local engine, four coaching calls a week,

3,800 members as well. And also with Hermes agent, we've built a lot of custom stuff that's really cool. So, for example, we've got the chat section here where we can switch between profiles. We got the talk station where we can talk to Hermes station directly. We got a voice activated version of Hermes over here and we can see our conversation history and everything that we've built with it. And then we also have Hermes Oracle which pulls

in the latest news and we can create content around that directly here. So this pulls in the latest trended news every 24 hours. We have the studio for generating images, videos and voice. We have the camb board as well to orchestrate teams of agents and we just set up very recently mixture of agents which is a way to achieve like fable 5 level intelligence with Hermes agent and a combination of different models inside Hermes agent and

the outputs look really amazing. So feel free to get that link in the comments description or just go to the arprofer.com inside the community. This is my community focused on helping you save time and scale with AI automation. Inside the community, you can ask questions, get help and support um daily. I answer yours questions with video tutorials. You get access to all of my best new training. So, if you're a complete beginner, you can go from

beginner to expert over here. If you want to get the agent OS system with Homies plugged in, you can get that over here. We've got the last update. So, we update this daily. Get the zip file full guide and a video tutorial on how to set up. We've got new daily tutorials. So, for example, the mixture of expert setup that we just used. We've got a full guide on how to use it in a 30-day road

map right here. And then inside the calendar, you can jump a week coaching calls, share your screen, ask questions in real time, and inside the map, you can meet people in your local area who are building with AI agents like Hermes. So, feel free to get that link in the comments description or just go to the aripboard.com. That is four free ways to never pay for Hermes again and use it for free. Thanks for watching. By

the way, if you're watching this and you have any questions or anything like that, like, feel free to ask. That's what I'm here for. Happy to help you. for paperclip is for orchestrating teams of agents as a company. So if you look at the way that you'd usually use AI agents, you use them inside the chat or you get them to build something, but you have no way to orchestrate like multiple different agents working together and

to really organize what they do. So with the paperclip system that that we have inside the agent OS, we have a team of agents working together. We can see what they're building. We can get them building 24/7. We can actually see everything that they've created in one single place with the live builds. And they created all this for us. We don't create any of this stuff and then they actually message us inside the inbox and then

we can see everything they built as well, right? So it's a it's a way to basically create like a zero human company. That's what we're looking at here. Whereas for example, if you have like a group chat with agents, it's cool for coming up with ideas, but it's not that organized, right? If you have a pipeline here for generating ideas, that's great, but how do you know what the organization structure is? How do you see where

the agents are? And so that's the problem that Paperclip solves is that it creates a team of agents working together with an actual hierarchical structure. Isn't that what a few of the other pages do already? Um, I would say, you know, show me one other page where you've got an organization structured like this. I don't see I don't know of any other system that has like a an org chart of AI agents working together or like

a a setup where the agents can inbox you proactively. So it's just bridge mind. Um I don't know. I don't know. I haven't really I mean I've seen his stuff. He's his content is really good but I've not seen the systems that he set up. But I know he's working on like a agent like that, right? So Hermes just released Hermes mixture of agents which is a powerful way to basically have a council of agents working

together to get better outputs. And this is obviously based on like the restrictions with GPT 5.6, Fable 5 being taken down. It's like, okay, how can we achieve better levels of intelligence using systems where agents work together and then you fuse the ideas to get the best possible outputs. And from what I've tested so far, this is a really powerful system. And I'm going to show you exactly what we've built with it, how it works, etc.

Um, how to use it and also how it performs on Goldie Bench versus everything else. So let's test this out and see how it works. Now if you want to okay how does this whole system work together. So basically I call this the Hermes council engine. So you have a panel of Frontier models merged by a chair and I actually put it through my whole 42 task leaderboard. So I've tested it with 42 tasks. You can

check them out on Goldie Bench as you can see right here for Hermes mixture of agents and actually ranked number two on the entire board pretty much above every single model out there. The the only one that's beating it right now is Fusion and also Grock uh sorry Opus 4.8 was outperformed by a long way and I'll show you the comparisons in a second. So if you're wondering how does it perform on the benchmarks is doing

pretty well and I'll show you the examples in a minute. Now if you look at the quote from news research on this who created Hermes agent they said Hermes agent now exposes mixture of agent presets as virtual models capabilities beyond the publicly available frontier. So if you're wondering okay what is this? Well basically what we've done is we set up a council engine which is a new tab built inside Hermes agent inside my agent operating system.

This is the setup right here. And so it's a new tab we built inside the agent operating system. And basically you can pick a few frontier models as you can see right here. Any providers mixed together. I'm running Opus 4.8 and GPT 5.5. And then when you ask a question, both models answer it privately at the same time. Then a third model, the chair, reads both answers, judges them, and writes one final answer that's better than

either. And we can see the actual outputs here in terms of the stuff created. So the panel of models is the council, the chair is the aggregator. You have one question that you put in and one better answer out. So news research shipped this idea as a mixture of agents. I wired it into a tab, gave it a workspace and put my own leaderboard on it to see if it actually holds up in terms of performance.

And basically the way this works is like a panel of experts would be one genius. So if you picture one brilliant person answering a hard question alone well if you now picture a panel each expert writes her own take privately a sharp chair reads all of them and gives you the best combined answer. So the panel would win pretty much almost every single time. So you have one prompt opus 4.8 for example and GT 5.5 work

together and then you get a chair that works and builds it and and fuses it all together. Now, in terms of the performance, you know, you might be wondering, okay, how does this hold up? So, if we have a look at Hermes mixture of agents here on the leaderboard, it's coming at number two, just below fusion. And if we pull up the answers here, you can just test yourself out yourself, right? So, if if you're thinking,

ah, this is not that good or it doesn't look that great, etc., just have a look on on this website and see what you think for yourself, right? Um, you you can make your own mind up or you can test it yourself. That's really why I created the Goldie Bench is because we wanted to test all these models on example prompts rather than just, you know, listen to some theoretical benchmark in a lab somewhere about uh,

you know, an AI model we can't even use yet. So that's why we created this system. And you can see here, for example, the stuff we've created is super nice. Like looks pretty cool. It's very visual. We've tested out on loads of different stuff. So, for example, like this is called the Dragon Realm, which is a an open- source project, as you can see, which is pretty cool. Looks super nice. Um, the outputs are super nice

and I I'll show you how that compares against like, for example, Opus 4.8 in a second and also Fusion in terms of the outputs, but it looks really cool. Pretty much everything that we built works really nicely straight out of the box as well. Here's another example. So, this is basically like a 3D version of Skyrim, like an open world game. And you can see like it it looks absolutely amazing for something that's just like one

shot. You know, you give it one prompt, you get two answers fused out. It looks absolutely great. So, we tested it on loads of different benchmarks. As you can see right here, performed very well on pretty much all the tests. And if we compare it versus, for example, Fusion, let's see what we got back. So we can see head-to-head how they perform side by side. So for example, this is the output from Fusion, which looks pretty

cool. This is the output from Hermes. So it's more like 2D, but it still looks super nice. This is a game called The Dragon Realm. This is the outputs from Fusion, which again is number one on the leaderboards. And if we compare that versus Hermy's Agent, in some ways, this actually looks nicer, but they they've both creating like great outputs. Now, if you're wondering, okay, how does that compare versus something like Claude Opus 4.8? Let's have

a look here side by side. So, Claude Lopus 4.8 versus Hermes mixture of agents. So, this is the dragon game we created with Hermes Mixure of agents. Looks super cool. This is the one from Opus 4.8. This is the output from Opus 4.8 on like a Neon City simulation. Here's the one from Hermes. And this one actually looks nicer, I would say. And it's a bit more like you can actually move around with it. It feels

less linear, if that makes sense. And here's a racing game side by side. So, this one is from Hermes. Pretty cool. Pretty nice to navigate. This is the one from Opus 4.8, which just feels like a little bit buggy. See how it's kind of sideways when you're using it? Doesn't quite make sense. And also, the graphics in the background don't look quite as good. This is a 3D racer. So this is this is with Opus 4.8

is pretty much broken. Doesn't quite work as well. This is with Hermes agent using a combination of chat GBT and Opus 4.8 together. All right. Much better outputs, way less buggy, more fun to play, more useful, etc. So overall, like it's performing really well. I actually created this benchmark after Fable 5 came out, so I can't tell you whether it's better than Fable 5, but as soon as that gets restored, we'll test out. Same with GPT

5.6. And right now at the top of the leaderboards, Fusion and Mixture of E agents are the two best features I've seen to create the best quality outputs. And you can test it for yourself. Like all the stuff that I've created with this is just hosted on that website directly. Now the thing that I would say about all of this you know whether you're using Hermes agent whether you're using for example Claude whether you're waiting for

Fable 5 to come back is like stop chasing the model build the system right everyone is waiting on the next model the next Opus the next GPT but if you look at what just happened on the benchmarks a mix of today's models already beats everything else right so you don't need a new release you don't need to worry about gated access with this you can just use a smarter way based on what's already out there and

the model is a part you can swap. The system is the thing that you own and the time it makes is huge, right? You got Fable 5 in preview, you got GT5.6 delays, you got the next Frontier models that are not actually available right now to the public. And so the winning move is not waiting. It's squeezing more out of the models you already have by combining them. So for example, if you check out the agent

OS system that we have over here, we have the mixture set up right here. And this is so much better than using the terminal because you can see everything that you've created inside the workspace. You can switch the models based on what you want to use. You can then ask the panel anything and run the panel whenever you want. Right? So you just click run the panel like so. It's the same for example if you're using

Fusion which is very similar process. You can use it exactly the same way. So you can chat with over here and then see everything that you've built on the side panel here. And the same for example Sakana Fugu another alternative from Japan that does a similar sort of thing didn't quite perform as well as Hermes mixture of agents but this is a similar sort of process where you can compare them side by side you can build

stuff you can chat with it and then you can get the workspace over here right so however you want to use these models whichever system you use you can build them all in to a process like this and then if we want to achieve like better levels of intelligence from the models we already have access to this is how you do And it's super simple and easy with a system like the HOS. And all these models

are swappable parts, right? You want to think of those as interchangeable, but really it's a system that stays. Cuz for example, these models over here are going to change in 3 to 6 months anyway or even next month or next week, but the actual engine you use to run all this, that's going to stay the same. Now if you want to get my agent operating system with the council engine built in the council engine lives inside

the agent operating system I built the same stack the boardroom runs every single day. So you get the full agent operating system Hermes Claude open claw wired into one dashboard or shared memory. You get the mixture tab with your own model council plus fusion and sakana panels. You get weekly coaching calls, daily tutorials and a 30-day road mapap to set it all up and token optimization tutorials. So it runs cheap as well. So if you need

to run this on cheap or free models you can as well. So if you look at the system as well, like if you're using, for example, mixture of agents, the great thing about this is, you know, for everyone else, like the other 99% who's messing around chasing models, the problem with that is like they're always behind. They pick one model and hope it's the best one. They hit that model ceiling and stop there. They wait months

for the next release. They beg for access to the gated frontier models. And you know, the APIs when they do get released for this stuff is super expensive. With this system that I'm showing you, you know, you get the best models today. You mix several models into a virtual council. You beat every single model on its own. You don't need to wait for anything. You can just squeeze from what's already here. You go beyond the public

frontier without any extra access not uh needed and you can swap everything in when a cheaper or better model lands, right? You can switch and change and chop this up however you want. Now, some people say, you know, the the best model wins. I just need to pick the right one. But actually, the two best things on my whole leaderboard are systems. So, if you look at Goldie Bench and you look at the top two performing

models, Fusion and Hermes, Mixture of Agents, they're both panels. They're they're not individual models, they're systems. And that's what seems to give everyone the edge. It's not about the models you use, it's about the systems you create. Other people say, well, you know, stacking models, that's lab stuff only, too complex for me. But actually, if you've just got a simple system like this where you can type anything you want and it's just one click away, that's

way easier than, for example, having Claude in one tab, chat GPT in another, blah blah blah. Also, some people say, well, I'll wait for GP 516 or I'll wait for Fable 5. But none of those have a confirmed release date. We don't know if they're coming out, if they're coming out at all. It might be a case that Frontier Technology in the future doesn't come out to the public, that nobody gets access to anything uh that's

the latest stuff anymore. We might get access to only models that are one year old, for example. That might be the future that we're living in. So, you just have to bear that in mind as well. And you might be saying as well, this might be technical to sell, but you can see so many tutorials and uh so many wins and testimonies. We've got 191 pages of wins and testimonials from people using our agent operating system

from the profit boardroom. So I know like if I can do it and I'm non technical and all these people can do it as well then I know that we can all win and learn together right people are getting amazing stuff with this. So overall what have you learned? You've learned that Hermes mixture of agents is really really good. You've also learned how to stop guessing. You've got proof. I've shown you the proof and if you

want to check it out for yourself it's on Goldie Bench. I've shown you sideby-side comparisons versus Opus. And I've also shown you sideby-side comparisons versus Fusion as well. And we've also learned today that it's not about the model. I don't think you need to care about the model anymore. I think what you need to do is just get great systems. And bear in mind, like most things you want to automate, like for example, we have a

system where we can automate videos and it it plugs in the avatar, it does the research, creates a script, it creates a voiceover and plugs it all together. This whole system doesn't need a frontier model. We don't need GPT 51.6 to run this. But it creates better outputs than 99% of people. The same with our SEO deployment systems. Like we can easily automate keyword research from our Google search console or from open SEO. We can generate

the content, deploy it to our website. We don't need frontier models for that. So I think like most things that you're working on don't actually need the best models. They just need the best systems. And that's what you get inside the ancient OS. Now, if you want to get that from me, link in the comments description or just go to the aiprofitboarding.com. And inside the community, you can ask questions. I personally answer all of the questions

inside here every single day with a video tutorial. Inside the classroom, you can get access to all of my best trainings with new daily updates. We've got a beginner to expo course over here if you're new to this. And we also have a new daily update section here. So, you can get the agent OS system here with video tutorial, the last update date, a zip file to install it, and also we add new tutorials. So, if

you want to learn more about how to use Mixer of agents exactly with a step-by-step road map and all the terminal commands and everything else, you can get that right here. Also got new tutorials on claude web design and everything else that comes out. You can also jump on weekly coaching calls, meet people in your local area who are using Hermes and other agents like this. And that's all inside the AR profit boardroom. Thanks for watching.

Uh, thanks so much. I really appreciate that, by the way. That's super kind. We'll keep that pinned. That's a nice one. Appreciate it. Thank you. Heat. Heat. Heat. Heat. Heat. Heat. Heat. Heat. Heat. Heat. I Hey, hey hey. Hey. Hey. Hey. Hey hey hey. Heat. Heat. Heat. Hallelujah. Falling. Fall. Hey. Hey. Heat. Heat. I feel Heat. Hey, Heat. Digital insanity. There we go. Fire. Hey Heat. Heat. Good news. Hey hey hey. Heat. Hey, Heat. Following following

Heat. Heat. Are you happy? Hello Azer. Welcome here. feel. Heat. Heat. Heat. Heat. I don't have the option to extend it for specific people. Heat up here. Everyone else. It's going to be It depends what you're trying to set up, but you can use the existing CLI you have or you can use free APIs. So, for example, if you're trying to set up Claude inside the agent OS, then you can ask Cord to set up the

chord CLI inside agent OS. Or if you wanted to use free models, you could use the local section and plug in free models that you have. Depends on the feature though. So for example, like Hermes Jarvis would use chat GPT realtime API. still rocking. There we go. Hey Jason, if you're struggling Same with that. I would recommend that instead you just simplify and go with one thing. But then when you get used to it, you can

always switch. If you have any errors though setting it up, like feel free to send me a screenshot and I can take a look or ask the agent that built it for you. That's fantastic. Feel free to share any screenshots or the prompts as well inside the community. We'd love to see it. That way I just show the work I'm already doing and it doesn't take up any extra energy cuz have to do that stuff anyway.

Okay, let me know how it goes and if you have any questions, feel free to reach out. Who we got here? Amry from South Africa, welcome here. And welcome from India too. Thanks for joining. Just building out some cool stuff with local agents using Claude. Heat. Heat. 8,000 steps. Not too bad. Not too bad today. Hopefully, we'll crank that up later. So today we're going to be looking at Quifos 9B which is uh a new local

model as you can see right here. It's come directly from it's available on O Lama the host and there's many different models. You can also see the context window here is 1 million token context window. I'll come on to that in a second, but essentially this is designed to be like a claude style creative reasoning model with a Quen 3.59B base. And so what you can do here is you can easily install it from OAM. If

you don't have OAM set up already, this is a way to run free models. You can use this terminal command. Once you've done that, make sure you have OAM running. Then you can run this terminal command and you can download the model and start running it directly. We've already plugged it into our agent OS system so that we can test it out. You see, for example, we've got it inside the build section here. I'll show you

what we've built with it in a second, but we can see all of our creations inside the local engine of our Aentic OS. So, I'm going to run you through what it means, how it works, what it is, why it's different, and this is basically a free claude style AI that runs on your own Mac. Now, is it like Opus 4.8 eight level? Probably not. Probably not. But is it decent and can actually create stuff locally?

Absolutely. So, Quifos 9B is a Claude Mos style creative and reasoning model that runs 100% on your machine, which means no cloud, no token cost, nothing leaving your computer. And we've already wired it into the Agent OS local engine, tested it out, and I'll show you exactly how it works. So, you can see the details here. So, it's it's pretty small as a model, only 5.6 GB when we ran this version. and it was free to

get from O Lama. You can also see some stuff that we created with this just for fun, you know, to test out, see how it works. So, we created like a little to-do list here. As you can see, we can add new tasks. We can clear everything that we've created. So, you can create like little mini apps. Uh this was like a digital clock that we created that runs locally. And this is like a little snake

game. Classic. Classic. So, we put it to work. We tested out actually works, which is pretty cool. And it's running privately. Here's a few more things. So, actually, one thing that surprised me genuinely was that it could create like a fairly nice design. I've seen I've seen like Frontier models create worse stuff than this for the landing page, which was pretty nice. It created some stuff that was a little bit glitchy, but it even created like

this calculator that seems to work as well. So, we tested it with like six apps, all three running it with OAMA. You can see how it works inside the terminal right here as well. And there's two models that I've seen recently that have been interesting for local, which is quos and also or 9 or 9B as well, which is something we tested out. From what I can see so far, I would genuinely say that probably or

1.0 is outperforming it. I've got another tutorial on that, but so far it's pretty good. Now, if you're wondering, okay, what's the build speed like? So, you can see the Quos was um a lot slower. Sorry, it had more tokens per second. on short and build designs. It's a lot smaller as a task. It can build working apps and the context window is insane for this which is 1 million tokens and also you know onif and

quif are both local free models. Now both of them are great you know but for the same work quas actually came back twice as fast which is pretty cool. Now you might be wondering, okay, why is it called that? So it's a stack of four deliberate moves layered on top of an open quen 3.5 9b base. That's why a model small enough to to run on a laptop. I mean 5 GB is really really small for

a local model. It can write and reason like something far bigger. So the way this works is like it's based on quen 3.59b which is a strong open 9 billion parameter base model. Then you got um something that's trained on like claude style reasoning and creative. From there it has a million token context window. So the context window has been stretched and it's also quantized with o lama and llama cpp so it can think like claude

locally. It's designed for creative stuff. big context window reasons and cause tools which is good for agentic h you can use it like so you can also see the benchmarks here as well also you might say okay million token context window that's amazing that is really the ceiling it's not going to run out it's not going to run like that out of the box so for example it can still run out of tokens on requests because

would load it with a much smaller window the token window was tiny when we tested it out actually. So depending on what memory you have, that's going to determine how many tokens this actually runs with. So the longer the window, the bigger the memory cost. So in the real world, you would pick a window that fits your RAM. And then you can see the different models here. So we use this model. There are some higher quality,

more heavy models here. as you can see and you've also got a near lossless state for Q8 as well. So there's lots of different models in terms of the size and everything else from Alama. So you can go with smaller which is lighter and faster but slightly less sharp or you can go with bigger which is closer to the original. So that's why we ran it in the middle. Honestly I run it on a Mac Studio.

You can see my settings over here and it does struggle with local models sometimes. So, I still don't think like local models are anywhere near Frontier level, but they're pretty fun to try out and they are getting better and better. So, what's great about this? It's free and it's private. It's pretty fast for its size. It's pretty light. Claude style brain agent ready, so it has native tool and function calling and it can actually build stuff.

What's not great about it? So, obviously, it's an IB, so it's not a frontier model. The million token context window doesn't actually happen in reality based on my personal setup. It can be a bit slow loading up and there's no built-in tools or memory or anything like that. It's just a model. Occasionally it gets a bit messy as well. When we tested it, sometimes it would work really well and then sometimes it would be a bit

messy. So, for example, like I was talking about the context window here, when we were testing out inside the chat, sometimes it wouldn't fully load or it just cut off as you can see right here. But then sometimes it would work perfectly as you can see right here. Right. So sometimes it does get cut off despite the large context window. So overall it's pretty cool. Cool. You know, it's worth checking out quos 9B 9 billion parameter

built on quen 3.5 clawed style thinking thinking model million token context window. You might say well it can't be any good but actually you've seen it run you've seen some of the stuff we've created. It's not bad. How it works is you ask it runs it. Quifos thinks. Um this all stays offline on your Mac. So if you don't even have internet you can still use this AI which is pretty cool as well. You also might

think like I need a massive setup for this, but it's only 5.6 GB. To put that in perspective, like if you're installing Gemma 12B, that's about 20 GB. So, it's about a quarter of the size of Gemma uh for 12B, for example. Bear in mind as well, you can plug this into whatever you want. So, whatever agent harness you want. So, you could use this, for example, locally via O Lama with Hermes agent as well. I

haven't tested that. I don't know if you'd get good responses, but it could be worth testing out and seeing for yourself what it's like. And if you want to get our local engine wired into a system, not just a a chat with the terminal, then you know a local model is powerful, but a local model running as the engine of your whole agent OS is pretty much unstoppable. So inside the air profit boardroom, you get our

agent operating system. The wise quos in as the local engine. So your agents can run on it by default. That means it's free, it's private, it can run with whatever you want. and the full agent operating system, including the local engine setup, the every prompt, the dashboard, four live coaching calls a week, daily tutorials, token optimization tutorials, etc., and a community of 3,800 builders. That's all inside the AI profit boardroom. Link in the comments description or

go to the profit.com. If you want to know what you get with the Agent OS, you can see some examples here. So, we have the local engine set up, which means that we can use local models. We can switch this in and switch this out. So every time I test a new model, I I switch a model and then we can then see the workspace here. So everything that we've built, we can preview everything that we've

created as well, which is pretty cool. We have free clawed code as well, which you can run with local models. And then for example, if you like agents like Hermes, we've got all these different styles of using Hermes. So we have Hermes Javis for example, Hermes Oracle built in for the latest news, a studio, we've got a workspace for CNF. created a lead generation tool as well for outreach. And then we also have mixture of agents

which is a brand new model from Hermes, a new system you can use to get like better outputs than Opus 4.8. It's pretty wild. So if you want to get that link in the comments description or go to the AI prof. This is my AR community for helping you save time and grow with AI automation. Lots of people inside there using local models inside the classroom. You get access to all of our best trainings and you

can get the agent OS from this section along with new daily video tutorials and guides based on what you want to use. Inside the calendar, you can jump on weekly coaching calls, get help and support in real time. And inside the map, you can meet people in your local area who are building with AI agents like you. So feel free to get that link in the comments description or just go to the apiprof.com. Thanks for watching.

So today we're going to be testing out ornith 1.0 which is a new selfarning local model you can use. Scores pretty well on the benchmarks. You can see some examples right here. So, this is a new free local model you can use. You can see how it compares versus Opus 4.7 here and it's crushing on all the key benchmarks as you can see. Now, we've already plugged it into our agent operating system which you can see

right here and we can switch the model anytime that we want. So, we can also run this with Hermes too. So, if we go to Hermes, we set up a separate profile for on and it can work aically with Hermes as well, which is pretty cool. And you can get it for free on Lama. So, just make sure you have Lama installed and then you download Olama. Make sure you've got it open. And from here, you

can plug this into whatever you want, right? Um, and you can see how to to get the model. Now, if you're wondering how does this work, what is this doing, etc. So, basically, this is a powerful system where you can run a local model. It's it comes from a company called Deep Reinforced. They dropped on and it's an open coding model that scores pretty well on the benchmarks. If you want to see what we've built with

it, here's some examples of different creations that we've made. You see the full gallery here, and these are all different things that were built with on. So, it actually can build useful stuff, interesting things, uh, even apps and games and tools and that sort of thing, which is pretty cool and some nice visual stuff as well. Right. So, it actually works. You can plug it into Hermes as well. You've already seen inside the chat that we've

got Hermes or Nif plugged in and it seems to work pretty well. Now, the interesting thing about this as well is like you can have a a self-arning agent which is Hermes agent and then you've got a self-learning model which is on and then you've combined them together, right? So, you can kind of get the power of both just by using this system. If you're wondering how does it score on benchmarks, so it is outperforming Gemma

4 31B on the benchmarks here. And if we look how it compares versus flagship models, it's it's not doing too badly on there. I would test this stuff out yourself. I don't really pay much attention to benchmarks, but if you want to see the benchmark comparisons, you can see them right there. Now what's interesting about this is it's self-improving. This is a strategy. This is the official blog from Deep Reinforce. So it's a self-improving training framework

that jointly learns to solve tasks and to construct the scaffolds that guide those solutions. So it's building its own setup to get things done. So it'll roll out multiple solutions to solve the same problem as you can see. So you got on here, you give it the task, it starts working on building the system to build it out. Then it comes with three solutions to get that completed. And then from there it selfarns and improves for

next time you use it. So that rather than relying on like a fixed human design harness, it treats the scaffold itself as a learnable object that covolves with the policy. Right? Basically what that means is like it's self-improving and it figures stuff out itself. It's a lot more independent. Now this runs with RL which is reinforcement learning. And so the way this works is it gets a task, looks at what was previously used for similar tasks,

looks at the task description, then generates a solution roll out, and then that's repeated over training to give it a positive feedback loop so that it keeps improving over time. So if we look at the benchmarks here for example and we compare it versus quen 3.5 3.7 max even which is pretty impressive I can't run quen 3.7 max locally but if we look at the benchmarks here we don't need to right on terminal bench 2.1 or

n 1.0 O is outperforming Quen 3.7 Max. Same for Terminal Bench 2.1 on Claude Code SW bench verified like it's crushing on the benchmarks here. So if you're wondering, okay, how does the whole self reinforcement learning process work? So it writes its own instructions. That's the trick here. And it's a reason a 9B model that you can run on a laptop scores so well and better than models many times the size. So most AI coders work

inside a fixed recipe a human wrote. The recipe, the harness, tells the model how to solve a problem. Look here first. try this then try that etc and every model gets the same recipe but if that recipe is wrong for your problem the model is stuck with it whereas with this situation or would write its own recipe which means for each task it does two things first it writes a plan so how it's going to solve

the problem and then it follows that plan to write the actual code but when the result gets graded the score doesn't just teach a code it teaches the plan too so both halves get improved and then when that's trained over millions of times the model doesn't just get better at writing code. It gets better at writing the plans that get better coding itself. So the strategy improves itself. It's a self-inforced uh positive feedback loop. And deep

reinforce actually calls this self scaffolding. So basically it's a model that learns how to learn the task then solves it. And that's why it performs so well. So here's how it works, right? You get the task, it writes a plan, it writes a code, then it gets graded and then that feeds back into the system. Now, you might say this sounds technical to set up, but literally you just get a llama and you use uh one

prompt command to download it. So, it's pretty simple and easy. And also, you might say it's too slow, but it ran pretty well on my system as well. And if you want to see all the demos that we created with ONIF, we've actually built out a local leaderboard. So this is ranked across the tasks that I give it. And then we grade the the size, who runs it, the license, how many tasks, how it's scored across

each task. Obviously, it's not going to perform as well as Frontier models, but this is a local leaderboard and we've separated that just to make it fair. And then we also have a main leaderboard here for Frontier models as well. So, you can sort of compare the differences and see them side by side. So, everything that we've created with all of these local models, you can check out here as you can see, and you can see

what we've created and how we've built with it, etc. But if you actually look for example like this landing page pretty nice, pretty sleek when it loads up. Looks super nice. Outperforms some Frontier models on the same test. So he's it's pretty good. I would choose this over something new that came out for example like Gemma 12B that seem to be too slow for me to run and also Quifos which seems to break sometimes when I've

tested it locally. And that's a new model that just came out. Now also if you plug this into our agent operating system let me show you some of the systems we built for this. So first of all we have the local Hermes engine. This is a offline agent that you can talk to and this is driven by right. So inside Hermes and the section here we can switch to and then we can run Hermes agent locally

offline for free. We also have an agent cambban. So what we can do is we can actually have teams of local models like on or if working together inside a cambban board like this and orchestrate them as well. You can do the same inside paperclip too. Then we also have a local chat system. So if you go to the local section here, we can use local models, see what they built, save inside the workspace, preview it,

etc. But there's a lot of interesting things you could do with this around the systems that we've already created. And the benefits again are that they're free private fast to run, and local AI is getting better and better. You might as well think okay how does it compare against queros which is a new model if you're into local AI or is designed more for agentic coding right whereas for example quifos is more designed for like creative

writing and reasoning and olif is based on the gem 4 and quen 3.5 setup quif is based on quen 3.59b also big difference in context window here. But yeah, for on that's great for coding. That's what it's designed for. Quos is more for example for creative writing and that sort of thing. So if you want to get all of our systems for using this stuff, you can get it inside the AI profit boardroom link in the

comments description. We've got the full agent OS inside there with the local engine, the full Hermes setups and systems that we have right here and our mission control for managing all your agents in one place. plus a camb board where you can actually orchestrate teams of local AI agents. So feel free to get out inside our AI community link in the comments description or go to the air profitborn.com. Inside the community, you can ask questions and

I answer them personally every single day. Inside the classroom, you can actually get access to all of our best training. So if you're a complete beginner, you can get the beginner to expert training here. If you want new daily updates on the new advanced stuff, you can get that over here. And we also have the full agent operating system that's updated daily with a video tutorial, full step-by-step guide, and a zip file to get this installed.

Inside the calendar, you can drop a we coaching calls, get help support real time. Inside the map, you can actually meet people in your local area using AI agents like you. And this is all available inside the AI profum link in the comments description or go to the prof.com. Thanks for watching. Cheers. All right, peeps. I think I've answered every single question, every single comment, every single question inside the community. showed you six different new things

with AI. That is it from me today. Appreciate everyone who tuned in. I'll see you on the next one. Cheers. Bye-bye. Patrice says, "Best community on the web right now. Very active place." Thank you so much, Patrice. Appreciate you tuning in. I'll see you on the next one.