Qwen 3.8 Max Full COURSE 1 HOUR (Build & Automate Anything)

[00:00:00]Quen 3.8 full free 1h hour course. Today I'm going to show you exactly how to use Quen 3.8. Some of the best use cases for it, what we've created with it, how powerful it is, how it performs versus other models. Now this is an anticipation of Quen 3.8. So Quen 3.8 is launching. This comes directly from Alibaba and I'll come on to the history of the models and how that's worked in a second. Um, but also this is a massive 2.4 4 trillion parameter model. The model is continuously evolving during the preview mode that I am making this video during, so it'll probably be even better by the time you're watching this.

[00:00:38]And they actually believe it's one of the most powerful models available today, compatible to leading Frontier AR models, second only to Fable 5. But the difference is that Fable 5 is not really readily available. Uh you can only use it on like 50% of your subscription with Quen. And also this is not a open source model whereas for example Alibaba's Quen 3.8 is open source and it is also from China and you can also get free access.

[00:01:05]So there's a lot of benefits to this.

[00:01:06]Also bear in mind that Quinn itself can make images but actually Fable 5 can't.

[00:01:11]So let's get straight into exactly how to use it, how to get the most out of this and how it works. And just to be clear here, we did have a further update on Quen 3.8. So they said during preview quen 3.8 is getting better by day. The latest version is live now. So is improving. It's getting better every single day. If you use the model yesterday for example, it wouldn't be as good as today. And I would expect that to continuously evolve um you know over the next course of the year really. And we're going to get straight into how to use this. Now you can use this inside Quen Studio which is available at quen.ai.

[00:01:50]AI. So chat.quen.ai.

[00:01:52]Just make sure that you select Quen Max Preview. You actually do not need to log in to use it. If you do want to log in, then you can just sign up for a free account over here. We can sign in with Google, for example. The benefit being, of course, that you can store your history and your settings if you're logged in and you can get a free account.

[00:02:13]And this is what it's going to look like once you're in. Now, what you'll see from here is you have a few different options. So, you got the community tab, and we'll run through all of these today. You've got the coder tab, which is also pretty interesting. Here you have from here, the new chat section, and search chat.

[00:02:30]So, if you want to find a previous chat you've been working on, you can search through there. We also have projects and also all chats are here. So, if we flick between these as well, you can see we've got images, we got videos, we've got reports, webde, dev, podcast, etc. So if we go over to new chat now, we're then going to select Coen 3.8 mass preview.

[00:02:50]Now what you also see is there's an option for model comparison. So you can actually compare your models and you can check out older models over here. So this is something that's evolved a lot over the years. You got Quen 3, Omni Flash, VL, KOD, etc. Uh all the way up to the current day which is 3.8 Max preview.

[00:03:09]And I will say the latest version Quen 3.8 8 Max is by far the best by a country mile. So over here we can switch over to dark mode just to make it look a bit nicer inside our settings. So you just go to uh your name and your profile here, click on settings and then you can switch the colors. You can change what language you want it in. You can change the voice. So it does have voice features as well. On the interface we can go with the default settings there.

[00:03:35]On the models you can actually see the details of each model.

[00:03:39]So Quen 3. Amax preview is a million token context window which is pretty good. And its modality is text, image and video.

[00:03:48]You can also import chats or export chats. So if you want to import or export to another account or to another model, you can. On personalization, you can select memories. So this just allows you to save memory so that the AI understands information about you and remember you and that sort of thing. You can also customize it. So you can change your settings over here in terms of who you want it to call you. So we can put for example Julian inside the settings here and then you can add information about who you are, what you do, what you use in Quen 4 and it will become more personalized based on what you recommend here. Now also you can see here that it's got all sorts of great features. So image search, it's got web search, image generation, code interpreter, like some really good tools right there. But I also like the fact that it's super simple and easy to use. And then if we go back to the settings now and we click on new chat, we can start getting into the magic of using this. Now, if you just want to use the basics of Quen 3.8, just make sure you have that selected.

[00:04:41]You can see when we click on new chat, it actually goes back to 3.7. So, you want to make sure you reselect 3.8.

[00:04:48]And then from here, we can start building. So, if I say, for example, build a beautiful website for an SEO agency. We can hit command and enter, and then it'll start working right there. You also notice that Quen 3.8 itself is only thinking. So, you only get thinking mode here. And you can also talk to it. So you can record your voice and then it'll actually listen to your voice and your recordings here. So if you don't want to manually type, you can actually just switch to record voice. And I will say when you're using Quen 3.8, it is quite slow to use. Just something to be aware of there. Now what you can also do here is you can have multiple different chats working at the same time. I would actually recommend that because it does take a while for Quen 3.8 to respond sometimes.

[00:05:32]So you can see here that whilst I'm talking to you, it took about a minute for that to reply. And sometimes it's going to be slower depending on the peak usage. So for example, if you're trying to code at the same time like the whole of the US is online and they're all trying to get access to a Frontier model when it's free at the same sort of hours, then just bear in mind like it may be slower during that time. So if you look for example at kimmy.com which is another free open- source frontier model Kimmy uh K3 particularly when you're using that like sometimes it's limited on the free plan during peak usage hours. So just be aware of that as well. So that is coding out and we'll come back to that in a second. What you're also going to see here is we have the coder section of Quent. So let's have a look and click on this and then we can connect our GitHub account to get started on coding like so. And now we have access to quen coder. We just need to add an environment and you can plug in the details of your GitHub. I actually don't use GitHub that much. Most of the stuff I create isn't open source. That's why if you come back to Quen Studio now, you can see we're getting access. You might also say, are there any other ways to access Quen? So it's free on chat.quen.ai.

[00:06:44]On KOD, I think they have a free trial.

[00:06:47]So you can see here that basically you actually get started back if you sign up for a new free account on Koda.

[00:06:54]You can get access to it and then you can use Quen 3.8 through there which I'll come on to in a second. This is like another coding platform and you can see that when we come back to the chat inside Quen, it's going to start coding out. Now one thing to bear in mind is like it's still outputting the HTML for our project. So it's not ready to preview yet. So if you see this just coding for a while, you can come back to it later. And you could also have multiple tabs open at the same time.

[00:07:18]That is a more productive way to get the most out of this because for example, if you're coding on this, you don't want to just be sitting there for like five minutes. You want to code out multiple projects in separate tabs so you can get more stuff done in the same amount of time and you can be more patient.

[00:07:34]You can also rename the chats whilst you're using this. So if we go over here, we can click on rename and then we could put for example SEO agency website right there. Hit confirm and that's going to start working. So Quen coder I'm not going to use so much because it's more designed for GitHub which I don't really use to be honest with you unless I'm using someone else's project.

[00:07:55]Now if we go back to Quen Studio here, you'll also notice that we can generate images. So if we click on max preview, if you're using Quen for images, there's actually no image option if you're on Quen 3.8 Max Preview. So, what you can do is switch over to 3.7 plus instead and then you can create the images or the videos which we'll come on to in a second. This is still coming out. So, let's create an image here. We'll just say create a image of a cat eating a sandwich and that's going to generate the image for us. We can also change like the aspect ratio and everything else. It's not actually using quen 3.8 to generate the image just so that you understand that it's actually using a separate image model. Now, at the same time, we can also open up another chat. So, you can have multiple tabs here. So, if we go to another tab over here, we can actually click on the create video here.

[00:08:44]Now, some people watching this going to be like, well, if you're coding loads of stuff at the same time, mate, what what's the point of teaching a course?

[00:08:50]But I do think this is the best way to use it. So, the thing that you're going to find is like you can't just code one thing at a time with Quen because it will take too long, right? And I'll show you like we coded out 50 different projects. I'll show you some of the amazing stuff he created in a second.

[00:09:05]But the thing to note here is like this is the way to use any AI tools. You have it running in the background autonomously on different projects. You might have one that's coding out. You might have one that's creating a a cat eating a sandwich, you know, because why not? Why not? And then you might have one that's creating a video. So don't worry about flicking between tabs. Don't worry about, for example, staying focused on one thing. The way to get the most out of Quen in the format that it's currently in is to use it like this. So that you can have agents working in parallel on multiple different projects and then instead of waiting for one thing to be done each time, you can code out like 10 different projects at once and you get 10 times more done in the same amount of time. To me, that is a smarter way to learn and a smarter way to build.

[00:09:52]So if we have a look here, this is the quality of the image. Um, would I say it's as good as something that's an alternative, for example, like Nana Banana? No. But at the same time, this is free to use. You can generate images for free with Qu, which you just can't do with most other models. So, if you look at this, for example, it does look a little bit weird with the cheese around the whiskers. Um, apart from that, it's a pretty good image. Let's test out the video generation now. So, we go over to Quenz Studio here. We'll create a video. And again, you can change the aspect ratio. This is the size and the dimensions of the video. So for example, you switch to 1:1, 3 to 4, 4 to 3, 16 to 9, 9 to 16. If you don't want to create the video, you can just X off it. If you do want to switch between video and images, you can do that over here. So we can hit enter like so.

[00:10:41]Next up, let's click on the plus sign and we'll test out the deep research. So here you can switch between advanced and normal. Now with deep research, deep research is not just like a Google search. is like you can go really deep on a report and get lots of information back. So for example here we could click on advanced and then we could say create an amazing research report about Quen 3.8 its history how powerful is etc. And actually, because I want it to be faster and I want to show you the outputs, I'm going to go with normal. But you could go with advanced. Usually with advanced, you might be looking at something like I think about 20 to 30 minutes typically.

[00:11:20]Whereas with this, you might be looking at like 5 to 10 minutes maximum. And you can see here they actually ask you a bunch of questions. And just to be clear and recap, we've got the website over here. We have the image generation over here. We have the video being generated.

[00:11:33]And you can see the percentage that's been generated. And then on one more tab, we have the deep research report.

[00:11:41]Now again, if it's asking you loads of questions, you probably don't want to type for ages because that will take up a lot of energy and time. You can just click on record voice here and use it.

[00:11:50]Or you can use something like whisper flow. That's typically what I use to to just quickly give it a voice note. So if you look at these questions, it's like, okay, I want you to look at the details of the model, how it performs versus previous models and also how it performs versus Frontier models. also look at the historical overview in terms of the evolution of Quinn over its previous models since it began as an AI company from Alibaba and then also the intended audience for this report is someone who really just wants to understand from an overview how Quinn works what it means and how it's useful.

[00:12:25]So, we've now got that. We're going to hit enter like so. Some use cases for videos might be like Facebook ads, images, social media, that sort of thing. And then we've got the website being built up here. So, now we're going to click on copy. Now, you will notice that you cannot preview the code that you've created inside the chat. Kind of annoying, right? Kind of annoying. So, you might be wondering, okay, well, how do you how do you use that instead? So, what you want to do if you're coding out a website or something that you actually want to preview, you can select webdev.

[00:12:56]So, click on the plus sign and click on webdev and then from there you can actually preview and I'll run through all these in a second. Now, this is a good time to show you what it can build and how powerful it is. I've tested loads of different models on Goldie Bench and you can see the leaderboard right here. So, for example, we got fusion. We got Hermes of agents. If you don't know what they are, no problem.

[00:13:16]Hermes, a miser of agents of fusion are both like panels of models. So they're multiple different AI working together.

[00:13:22]Then you have GT 5.6 Soul speaks for itself. Claude Fable 5, the latest models as I'm recording this. And then we have Quen 3.8. And based on 45 tasks, Quen 3.8 is performed in the top five of all the models. Now bear in mind, it is in the top five. It is the only open-source model that you can get access to. No other model is open source which means that it's not shared the details of what it is or how the code works etc is not shared online. Um it also means that you can run versions of Quen locally. So a lot of people who have a good setup like a DJS Spark for example they love to run Quen 3.6 locally. It's one of the best models you can use for doing that. And you can see some cool stuff that's built here. So this is looking pretty cool. This is also like a open world game. So it's it's not bad at all. The the only thing I'll say with Quen 3.8 A from what I've tested is that you need to go back and forth for it a bit. It's not quite one of those models. It's not like Fable 5 where it just gets it right every single time. It's like you need to go back and forth it, get a bit of feedback, and then it will come back to you here. So, this is kind of like a Doom example here. Doom style game as you can see.

[00:14:29]Um, yeah. And the reason that we create games and test those out with this is because it usually gives you a good idea of like how it performs when it comes to coding, when it comes to graphics, when it comes to physics, when it comes to reasoning. You know, these are all the elements that required for a good 3D game, which you would then apply to the real world. The difference is that it gives you a very visual feel for how it performs. Let's test out another one.

[00:14:53]So, you can see like it creates some pretty nice stuff, right? It's pretty cool. definitely right up there with Fable 5 and GP 5.6. So, and bear in mind like really Kimmy and Quend 3.8 are the first two open source models to really challenge us and to be right up there with these models. So, that's impressive in itself. Here's another example. This is kind of like a open world game. As you can see, super nice like the UI and everything else.

[00:15:22]And then this was pretty fun to make. And I just want to show you these examples because you know this is how you go from just like a a bit of code or a prompt on a page to creating something amazing, right? You just have to give it the details and um all the prompts are available on Goldbie bench as well. And this it can actually create videos for you as well. So what we did here is we actually got coder qoda.com.

[00:15:48]Um, we taught via Cyod uh, Quen 3.8, we taught it the skill of something called Remotion. And Remotion is an open source skill that can help you create videos.

[00:15:57]So, it can create videos like this and code them out, and they look really beautiful. So, yeah, we've tested it on so many different things, like 45 different tasks, and it's created lots of awesome stuff. Here's an example of a website, a landing page that's built.

[00:16:11]So, if you're wondering, okay, what can you do with this? Everything, mate.

[00:16:14]Everything. videos, images, custom animations, uh games, apps, whatever you want to build, you can build with it. It's right up there with the Frontier models.

[00:16:25]That's the thing to note here. So, if we did accidentally create a HTML code like this, we can't preview it. You can just go to something like this. As you can see, then we can paste in the code inside the HTML over here. Paste that in and preview it. And this is bare in mind. It's just like a a oneshot website that we created, but it looks pretty nice. Um, it's got it's it's quite a nice website that's designed and we literally just gave it one prompt. So, bear in mind like this is free. This is open source. It's available to you. It's available to me. You can generate images. You can generate videos. You can code with it for free. You can create amazing games. Like it is a very powerful tool and very underrated. And this is really the first time I think anyone's taken Quen that seriously apart from local. Local people are absolutely loving Quen 3.6, let alone 3.8 Max preview. Also, it's not been announced when this will be released in terms of open weights. So, he is an open source model, but the open weights haven't been released on Hugging Face yet. However, I would say over the next couple of weeks, you should expect this to come out and drop. So, that's pretty exciting itself.

[00:17:30]Now, if we come back over to webdev over here, you'd actually see some examples of what people have used. So, for example, we could like write a fruit e-commerce website. And you might think, okay, what are the best prompts to use this? What are the best prompts to get the most out of this? Honestly, you don't need to worry too much about prompting. Every prompt, even the games I showed you earlier, the prompts are super nice. Um, but they're simple.

[00:17:52]They're very, very simple. And we're at the point now where you don't really need to worry so much about prompts.

[00:17:57]Like if you have an idea, the AI is smart enough to really give you an output. Now, what you're going to notice when you're using webdev is that when you click the plus sign and you do webdev, as you can see here, it's going to give you the code and the preview.

[00:18:10]So, once this is completed, you will be able to preview the actual code inside the chat instead of having to go to another website to preview it over here, which is pretty cool. And again, you can run all this in parallel as you can see.

[00:18:21]Let's have a look at the video now.

[00:18:24]So, the video does have sound as well.

[00:18:26]I'm going to mute it because it won't sound very good on the replay. But if we have a look here, this is the video generated. Again, is it as good as something like Veo3? Probably not. But at the same time, it is good. I mean, like, you know, and bear in mind as well, one limitation of this is that this is only a 5-second preview. You might say, okay, well, how do you create a longer video? What I'd recommend is that you use something like Remotion.

[00:18:51]Remotion is a free open source skill and it created that video promo that I showed you earlier. and then you can create a much longer video. So you can literally go over to Quen or Koda, give it access to this. So you just paste in the link to this GitHub for reotion and then you ask it, hey, can you create a video based on reotion and then you just go back and forth until you get the outputs you want. That's basically how it works. So now we've got the deep research set up here. And by the way, we have covered a lot already. We've covered Quen Studio and how to code inside there. We've covered how to make an image of a cat eating a sandwich. AK, how to make uh images. We've created a full deep research report, a video, and a website. And it's all looking pretty good. So, let's have a look at the output from Quen here in terms of the deep research report. What you're going to see is that on the left hand side, you have the full chat with the history, right? But when it gives you the output of the research, and it only took 6 minutes for the quick research, it's pretty good. When you look at this and we scroll down, it's not very nice to view in the chat here. So, what I'd recommend is that you resize it and you view the report fully like that, right?

[00:20:02]Looks a lot nicer, much easier to do.

[00:20:04]And you'll see how in my library now, we've got a full history of all the stuff we've created, which is great as well. Now, we can also click on create and then webdev or podcast. So, we can actually create a podcast from this. And then also over here we could download the research report either in markdown or PDF. So if you just want to preview it straight away you can get the PDF and then you can open it like so. And then if we open that up we can check out the full research report.

[00:20:29]Now when you look at this research report unless you advise it other ways the report is going to be kind of like academic wording. So it's not that easy to read but again you can ask Quinn to simplify it. Now, also something that's really useful to note here is the dates of of the evolution of Quen because you might be wondering, okay, how does this all work? So, Quen 2 was released June 2024. Quen 3 was released April 2025. And since then, the shipping and the iterations of this model has improved massively. So, you can see for example, this Quen 3.5 was released on June 2025 and then Quinn 3.8 was released on 2026. Quen 3.8 8 is by far the best model that's ever been released from Quen. And you can see the difference in parameter size here. So 0.5B was where it really began versus 2.4 trillion parameters over here as well. Now also if you don't have time to read the full research report, you can actually just go back to the chat here and you can say give me a breakdown of the full research report, how it works, etc. give me a breakdown of the findings from the research report in bullet points and just show me the most important details that would actually impress me. And then if you just want to get like a quick summary or a quick recap of how it's working, you can do it like so. Right? So the main points here are like 2.4 trillion parameter scoring pretty well on MMLU according to this big context window as well. However, what I will say and I just want to be 100% honest with you is like if you actually look at this and I was just reading a bit deeper into it. So I actually I looked through the information I was like this doesn't seem quite right because it was announced on July the 19th but I didn't see any information about this benchmark and then if we actually check the details of that benchmark you can see that it just uses like Instagram as a source which anyone could post to. So just be aware like it still hallucinates. It's still making stuff up. And actually when I challenged it inside the chat and asked for the links to the sources it number one doesn't understand that Quen 3.8 eight actually exists in the real world. This is default into quen 3.7. And then number two actually says that some of the text from the research report is entirely hypothetical and fabricated. So I would not use just want to warn you I don't use Quen for deep research as you can see right here. Not particularly useful and it made a lot of mistakes here. Also if you switch over to Quen 3.8 Max you can't select deep research.

[00:23:00]which is not available. If we go to a new chat here, we can also go to artifacts.

[00:23:06]And with artifacts, we can create like sort of previewable mini apps that you'd want to build. So, for example, here like okay, create a app for habit tracking and then it'll begin to build that. Now, at this point, we should be looking at what and how it performs versus other models. So, what I've done over here is I've built out a ton of projects using Quen 3.8.

[00:23:32]And you can see Quen 3.8 on the left hand side. We got GT5.6 Soul over here.

[00:23:37]We have Kim K3 in the bottom right. And we have Fable 5 as well. So, you can see here for example that if we look at all these tests like the way that it's been designed, the graphics, the gameplay, etc. is way nicer on Quen 3.8 8 versus any other model right now. Now, let's have a look for example at Skyrim. So, this is like a Skyrim style game as you can see. And if we have a look at how it performs versus the other models like it holds its own actually outperforms at least one other model here which is GT5.6 Soul. And if we look at the gameplay and everything else, like it looks nicer when it comes to this game too. And if we have a look, like it all seems to follow the same pattern like this one as well, super nice, nice 3D graphics, looks really cool, fun to play, etc. This is a neon blaster style game as well and the gameplay and everything else is really really cool. So, it can create great stuff and it does hold its own versus other models. It's kind of like a Zelda style game. Apart from the camera angle, it's probably created the best output over here. And let's just run through some more.

[00:24:45]So, as you can see, I think the only one I didn't really like from it was the flight simulator. As you can see, it's it's not that fun to play when it comes to the actual camera angle and everything else.

[00:24:56]This one blew me away. Like, if you look at the quality of this game versus all the others. So, like look at GT5.6 Soul versus Fable 5 versus Kim K3 and you you'll clearly see that Quen has created the best out here, the most fun output and everything else. Here's another one.

[00:25:15]I'd say actually on this one like the graphics are super basic, not that great. This is Twilight Veil looks super nice.

[00:25:23]This one is cool, too. And I think we're really at the point now with a lot of models where where China is holding its own, if not overtaking the US, which is a very interesting time. Maybe one that nobody imagined would get to, you know, like. And like look at this game. It's it's super fun, easy to play, and they're all like that pretty much, you know? They're all really, really good.

[00:25:46]Super nice. So, yeah, if you're wondering, okay, what can you do with it? Basically, anything. Now, also at this point, you might be saying, okay, I like Quinn, but it's not free. Some people, they always seem to comment on my videos saying that. But, as you can see here, there's not even an option to sign up. Like, you can use it for free as long as you use it inside the chat, as long as you use it on chat.

[00:26:09]Which is what we're doing here. You can do web design, you can do coding, you can create images, you can create videos. There's not even an option to get a subscription right now. Other people say, well, you know, Chinese models, they can't out compete or outperform US models. US is still the frontier, but as you can see right here, if we look at all the comparisons, Quen 3.8 is clearly winning. It's clearly better. Other people say, well, you can't code with it, blah blah blah. But you actually can code with it. We've coded out 45 different projects here.

[00:26:40]So, anything that you want to do with this, anything that you want to do with another model, you can do with Quen. And then other people will say, well, with Quen, it's it's not open source, but it is going open source very soon. So, they've basically announced it's open source. It's just like we're waiting for the open weights to be released. They always make each model open source. Now, also, you might be wondering, okay, what is KOD? How does that work? So, this is like a an IDE, like a desktop app. you can download and use as you can see. So we can get it for Mac, you can use it on Windows, however you prefer it to to be used and then you can get a free trial on here as well if you prefer to use it on desktop app. Bear in mind like after that is a subscription.

[00:27:21]So if you just want to keep it free then you're probably better off using the chat from Quen and Encoder you don't just get Quen. I think you do get multiple different models. There's another big benefit to Quen via a coder as well which is the CLI and basically what this allows you to do is like you can plug it into claude code.

[00:27:43]So if you have a look over here for example we have quen 3.8 plugged into claude code and we can build with it over here. All right so we can build with it. We can ask it to build stuff directly inside our agent operating system. This is inside a profit boarding by the way if you want to get my version or you can build it out yourself. And then for example, if we go over to the workspace here, everything that we've built with Quen 3.8 is inside this section. So we can see everything that we've created with Quen 3.8 over here.

[00:28:10]And then we can just install Coder like so. And also you might be wondering, okay, how does Quen 3.7 form versus Quen 3.8? So this is Quen 3.7 where we've built out a little habit tracker app. So you can use it for habits as well. as you can see right here. And then if we have a look, we've got the preview from Quen 3.8 over here. And here's something to notice. Like if you click on the preview section here and we try and use a habit tracker, we actually can't use it, right? So if we click on the days, for example, it's not really flickable.

[00:28:42]It's not previewable. We don't seem to be able to do anything with it. So it's like cool idea, but we actually can't do anything. Um, let's just click on save here. You see how it says have it saved?

[00:28:54]have it added, but you can't see it inside the app. So, this is Quen 3.7.

[00:28:59]And this was always why I didn't really use it for coding particularly because it was just kind of like it would create stuff, but that stuff didn't work. Not very useful. Whereas, if we go over to preview here, we can actually test out the habit tracker. Number one, it looks a lot cleaner. It's easier to read from Quen 3. Number two, it actually works.

[00:29:16]We can edit it. We can change it. We can delete stuff. And you can see it actually deletes. So it's a bit more like wow this is actually practical. And then you'll see the code over here if you just want to grab the code from the chat and you can deploy it somewhere else or you can go to the preview and view it over here. You can also view it on mobile. So it can create mobile apps as well as you can see here. And then we can click on deploy and we just plug in the name of the website. So click on deploy over here. It's going to give us this section like so. What that means essentially is like we can share the apps, the websites, the artifacts that we create with Quen with the world. So if you have a team and you create an amazing app on chat.quen, you can actually share it with the world for free, which is pretty cool as well. And it it creates stuff that actually works as as you can see right here, which is awesome. You if you wanted people to be able to log into something like this, you would actually connect something called Firebase. So if you're wondering okay like how do you create apps where people can log in and store their data and that sort of thing. Firebase is a great way to get started with that and you can just connect these. You can ask Quen like hey connect Firebase to this and it actually do it for you. And right now we are just coding out this fruit e-commerce website. We already got the 3.7 preview but again I want to show you the difference between 3.8 and 3.7. So we will come back to that in a second.

[00:30:29]And now we can sign in with coder as you can see.

[00:30:33]And this is what coder looks like. So this is another way that you can use Quen 3.8. If you've ever used a IDE before, this is something you're probably familiar with. If you've never used an ID before, no problem at all.

[00:30:48]Let me show you how to get started. So first thing we want to do is just open up a folder. And this way we're going to code locally. So we can click on new folder and we would just put something like for example AI profit bordering website. So we're going to create that like so. Open that up. Click on I trust the authors and then from here we can start building. Now what you're going to see is a section where we can select the model. Now you can go with the auto if you if you're not too fast but if you want to use Quen 3.8 specifically then you can just scroll down and select Quen 3.8 preview. So you can click on like so and then you can switch between modes.

[00:31:24]You got experts, you got agent and you got ask. Ask is like quick Q&A. So you can just have an instant back and forth.

[00:31:31]Agent is where you can actually plan and implement everything. So this would be the best if you're coding out a big project. And you also have experts.

[00:31:37]Experts is basically like where you have a team of AI agents working together. Uh and they work in parallel.

[00:31:44]However, that's more for like a big task or for example where you need like the best of the best outputs. If you're just building out a basic coding website or for example building out like an app or something like that, you can just go with the agent and then Quinn 3.8 preview. And if we say okay, build out a beautiful website for the AR profit boardroom. Here's an example of my existing website.

[00:32:04]Um, then we're going to go over to the arprofit.com, which is my AR community link in the comments description to go to the arrwarm.com.

[00:32:13]And then inside here, we can go over to coder. We'll paste in the HTML from my example website. That'll basically just train it up on like here's the copy is the information about my website. Here's the HTML and the branding we use. And we can also click on optimize input.

[00:32:31]That'll basically edit the prompt itself with AI. So you don't even need to prompt it yourself. You'll actually come up with the prompt. So we can optimize the input as you can see right here. And that will edit it. So it explains more without me having to spend more time or energy editing it. So you can see here like it fully edited the prompt and made it a lot more in-depth. And then we can also undo anything if you don't like it.

[00:32:53]You can check your usage over here.

[00:32:55]how much you've used and then we can click on send and we can begin to code out with this bad boy. So now what you can see is that it has this file with the code that is beginning to work on. Now the agents are actually going to just start coding for you. You don't really have to do too much here.

[00:33:16]It's pretty handsoff.

[00:33:19]So it says I have the content and you see here that it's just giving you updates. Says like right I have the content. The workspace is empty. So, I'll build a polished modern single page website from scratch using the existing HL you've given me as a foundation. And then you'll see here that it's got this to-do list. So, it's just going to mark off all the indiv individual things that it's doing step by step. So, for example, like creating a logo, previewing it, verifying it, creating the index html, etc. These are all like the basic files for your folder. Now, you might as well say, okay, what's the difference between chat.quen versus using an ID like coder? The difference is that with an IDE what you're creating number one can work across multiple files whereas for example if you have a look at chat.quen usually it's just like one single file one single piece of code like so. So if you want bigger more complex projects then you would go with coder.

[00:34:10]Additionally it's also coding locally.

[00:34:13]So that's a big difference is like you can code locally and you can save files directly to your laptop or to your computer and come back to them later.

[00:34:21]What it will also do is it's more optimized for like bigger projects. So, it's planning out the to-do list and it's more hands-off. It's more autonomous because it's just working in the background whilst you do something else. You could even have multiple projects being coded at the same time in parallel whilst you just go off and do something else. And then additionally, it's a CLI which means that you can plug it into an agent operating system or you can plug it into an AI agent like Hermes or OpenClaw if you're familiar with those and I'll show you that in a second. And then additionally, you can also change the model. So there's a few things to note here. When you're changing the model, you can switch between auto, ultimate, performance, efficient, and light. You can also switch to a custom model. So if you want to change the model yourself manually, you can. If you want to not use max preview, you can change it here. Or you can actually change the context window as well. So for example, if you are adding a lot of data to the conversation on whatever you're coding, then you could use a bigger context window. How do you change it? So you just click on the edit button here for quen 3.8 mass preview and you can switch to 400,000 tokens or a million tokens. And you can switch thinking on and off which you can't do on chat.coin.ai. So you get a lot more customization. It's a bit more of an advanced tool but it's better for building bigger better projects that are more interesting. Additionally, you might be thinking when you see it, why is the context window defaulted to 200,000 tokens? Because it will use less tokens basically. So, if you're on a a coda plan, you have a limited number of tokens on your subscription. It's not like the chat because it's it's creating bigger stuff and it uses more resources and so it's defaulting to a lower context window so that when you've got that context window, you're not using too many tokens and you don't like max out your subscription. Whereas, for example, if we switch over to like a million context window, it will use more tokens, but it will handle context more or handle more data better. And then you can see once it's done that it's actually going to ask you to reject or accept the changes it's made. So if you click on change files you can see here it's actually edited the index html which is a new file it has coded out for our website. So we can accept that like so and we can accept the changes that it's making. And you can see here that we've got the HTML ready to go. Now that is not the end of the project. So it's going to create more files for example like the CSS which is actually how it visually looks as a website. It's just going to do these step by step. As you can see here, the other thing that's a big difference between Koda and something like chat.quen is that you can add MCPS to Koda and it just feels a bit more natural. So any sort of IDE where you're using uh a system like this, you get a lot more customization. So for example, when I was talking about Superbase before, we could actually ask it to set up Superbase for us as a login for our project and it would just guide us through the steps step by step and we might have to log in or set up a API key along the way, but it's pretty chill to set up all this sort of stuff.

[00:37:18]Additionally, you'll see here that we can add multiple chats. So we can do things in parallel at the same time working on this folder here. So we could have, for example, another agent that's improving the HTML. Then we could have another chat here and this can actually reference different files and folders.

[00:37:35]So if we type in app, we can reference any other files inside our project, any folders inside our project, any rules inside our project which we'll come to in a second and any attachments as well.

[00:37:45]So it could be like okay check the HTML, make it more beautiful, check the CSS, remove any bugs, that sort of stuff. And then for example rules are basically like we can add rules here and what these rules mean essentially is like you are directing the agent in terms of what it can and cannot do. So it could be like okay every time you mention a free SEO strategy session for my agency link to julangoldie.com something like that. Okay. Now we can accept that vile and that has finished off the index html for us and it's just going to start working on the rest of us for us. Right. So, it's now working through that to-do list, which you can see over here, and you're good to go.

[00:38:22]Now, there's also something called Quest. So, you can open up Quest as well. You'll see this link inside KOD here. And again, you can switch to Quen 3.8 mouse preview. Quest is kind of similar to KOD. I mean, they basically work in the same way, but has a bit of a different UI. So, if you look at this, it doesn't quite feel like an ID in the same way. And also you could have like for example quest working on one project you could have coda working on another.

[00:38:47]That's something else you can do here.

[00:38:48]But yeah everything that you're doing here is previewable save locally etc. You can actually right click any files or folders you want to edit. So if we click on review with binder on the index html you'll see this page right here.

[00:39:00]Now we can actually preview this in a new tab on our Chrome like so. So we can drag that in and you can see here that we're beginning to preview the page that's been built so far. Now, this is early days. This is just an example of what it's created so far. It still needs to create the rest of the project, but you can see here it's using this as a base for the rest of the project. And that is very similar to our original HTML. And so, it's beginning to to code out and build all this out for us locally. And we can preview it before it's even ready like this. And you see here, it's beginning to add the rest of the stuff to the project as well. So, that's how you can code out projects with it locally. Now, let's talk about Hermes. Because to recap, we've covered how to use chat. And the chat is great for going back and forth. It's awesome for using images, creating videos, not so great for deep research. Wouldn't recommend that to be honest with you.

[00:39:52]Have to seeing the outputs. And then you've got coder. Coder is awesome for building out bigger projects. So that could be a mobile app, that could be a website, that could be something that you want to just build locally. And you can switch between the AI. And you also get a lot more control over how Quen 3.8 A is set up over there. Now what we also have is we have the agent operating system and this is a powerful system where you can for example have all your agents inside one place. So for example we have KOD plugged into our setup with agent OS. So we can switch between GP 5.6 or GM 5.2 or Kim K3 or Quen 3.8 inside our agent operating system and we can code out and see everything that we built over here.

[00:40:35]Right? This is really cool because now we can basically use Quen 3.8 whenever we want even though the API is not out because we're using the CLI from KOD which I was talking about earlier. And the great thing about a system like this is you can have all your agents working side by side. So for example, we got Claude, we got anti-gravity, we got Kodex, we got Kim code, GM 5.2.

[00:40:54]Everything is saved inside one place as well. So we can see the conversation history and everything that we built lately as well. Now let's talk about Hermes. If you're not familiar with Hermes, it is an open-source free AI agent that you can use. And the difference between say like an AI agent and a coder or for example the chat is that it can do a lot more cool stuff in the real world. And I'll show you some examples in a minute. But basically with Hermes agent, it's free, it's open source and you can plug Quinn 3.8 or any AI model into it. So if we have a look at our agent operating system, this is something we've custom built is that we have profiles for each of the new APIs that come out. So we have G, we have GRO 4.5, we have GT 5.6, we have Hermes cloud inside there. We even have, for example, Quen 3.8 already plugged in.

[00:41:37]Bear in mind that normally people would have to wait for an API to be able to to set that up inside Hermes, but we already got coder and plug that in as a CLI into Hermes agent as well as well as having it inside the coder over there.

[00:41:50]So what we can do from here is we can get it to agentic tasks. And the way that I would look at this is like Herpes is the doer is the hands. It can do stuff in the real world. So I'll show you an example in a second, but it can it can do more practical things. It can also, for example, look at web pages. It can learn skills.

[00:42:11]It can control your browser. It can control your computer and everything else. So it's very very agentic. It's not like a coding app where you're just building a project and it's not like a chat where you're just going back and forth for it. It's something more special. something more powerful. And so, for example, if we go inside the chat here and we're like, okay, for we give it a guide on information about us like this, and we just paste that in.

[00:42:35]So, we've done forward/ plug that in, and then we're going to hit send. And you can see Hermes agent working over here. Now, also, if we go over to Hermes here, we can also have different agents working in the background at the same time, different versions of Hermes working at the same time. So, for example, you could actually plug in Quen 3.8 into something like this voice agent. And this is Hermes Apollo, which is a custommade workflow we've built.

[00:43:00]And basically, we can speak to it in real time and it will speak back to us.

[00:43:04]Plus, it just listens to what we're saying. So, you can see here, for example, we have a full previous conversation of what we've built and when we've talked to Hermes agent.

[00:43:12]Really, really cool. So, if we say open up Obsidian, it'll actually open up programs on our computer for us.

[00:43:20]and it will reply to us as well. If we scroll down, we can see everything that we've built with Hermes Apollo as well.

[00:43:26]So, these are all projects that were built with our voice and we can chat with it. We can have it run in the background. We can see everything that we've done previously with it. We can get weekly or daily briefings with this.

[00:43:36]We can also have it in wall mode. Now, again, Hermes is just the hands, but if you want the brain for this, then what you can do is you plug in something like Quen 3.8 into it and it could run in the background. You could have it voice activated. It's listening real time etc. Now another option is you could use something like Heromy's Oracle. Heromies Oracle is basically putting in trending information from the web, trending news headlines and then from there we could create social media content through it.

[00:44:05]We could create SEO content for it or we can just stay up to date with the latest information. So if we look for example Claude Honeycomb Opus 5 leak that's looking pretty interesting and that is the most trending news today. And if we scroll down for example, we've got all these new trending news. And you can customize this to your industry, whatever you're interested in, etc. But the main point is we can easily automate our social media content, we can easily automate our SEO content just in one button. And we can preview what we've created previously. So as soon as, for example, Kim K topped the front end code arena, we actually created this blog about it and it's fully optimized for SEO, fully automated in one single click. And we literally just did that on an autopilot. This the other difference as well is like with for example Koda is coding that for you but you have to kind of babysit it. When you look for example at the chat you still again have to babysit it. But this system allows you to get three quen 3.8 and do schedule tasks. You can also connect Hermes to your phone and then you can control this agent that can do scheduled tasks and actually act in the real world and interact with all your websites or whatever you're working on in a way that you can't with anything else. So, for example, this is running on a 24-hour schedule. I didn't have to refresh it. I didn't have to ask it to do anything. It's just there as soon as I open it, and it's super chill. Final thing that I was going to show you was Hermes Astros. Hermes Asteros basically monitors our competition. We can see, for example, that we've got a watch list of keywords and people that we look at each day and then it pulls in the latest data on them and gives us content ideas based on what's just dropped and what's actually useful. And you see, for example, like some of these topics, they only came out 20 hours ago. Now, also, if we open this up, it'll give me the unique ideas for myself. So, I don't need to come up with the ideas like the system does it for me with Hermes agent.

[00:45:55]And then for example, we've also got this section where we can just plug it into our video agent and create a video based on that trending topic. We can for example create SEO content based on it.

[00:46:07]We could create infographics for free or podcast for free with no bookm as well.

[00:46:11]Now also we were looking at Quen 3.7 versus 3.8 Max preview. So let's have a look at the differences here. This is for a website this time instead of a habit tracker app. So this is the website created with Quen 3.7. As you can see, a lot of the links don't actually work. If we refresh the preview, just see what it was created a second ago. I mean, like the images is super nice as well. That is the benefit of using this is like it can create it own images and then add that to any coding projects that it's creating over here. And then if we have a look, this is 3.8 version. So this is the newer version. As you can see right here, we can actually open that up in full screen as well.

[00:46:50]The main point is here like it looks pretty nice. It's designed nicely. It actually works. It also doesn't look like your generic sort of website, right? Like this used a a very custom template there that I've not seen before with AI. So that's good as well. It's not like just looking the same uh like all these other AI slot websites out there. Additionally, we can also create slides with Quen, but again, that would be Quen 3.7.

[00:47:17]And there's some options for learning and traveling as well. And then we can also change the tools of it. So we can switch off switch on tools. This is back in the chat by the way. And something else that's quite useful here is project. So you can add a new project like so.

[00:47:32]We can click on a project. What that just means is like if you for example are working on let's say a brand and you want to categorize all of the chats in one place then I would choose a project instead. So it's something that you can come back to whenever you want to. And then you can train that project like so.

[00:47:51]So you can see, for example, if we start a new chat, it goes underneath the project here and it's kind of pinned at the top as well. So it's easier to get back to. So if you're just doing like a one-off chat about something totally random, then you can go over here. But if let's say for example this website was a long-term project. Well, we could actually put that inside a project system here and then we can go back and forth with it and see all of our conversations about that one particular chat in one place which is useful. Now with any projects inside the chats well you can pin them, rename them, clone them, archive them, share them, download them or move them as well. So you have quite a few options for the project to come back to. And we can also create presentations.

[00:48:30]So we could say create a beautiful presentation about Claude Fable 5 and the latest AI automation news. Hit enter like so and we can go that. You might also as well at this point be saying okay like do you use Quen Julian? Is this your number one AI model you recommend? Not for me personally. I still prefer Quen. Um sorry this is this is broken. Let's refresh that. That might just not work. Um but yeah for me personally I prefer to use claw desktop. I think that's just the way my brain is trained right now with Fable 5. I do think Fable 5 is just an absolute beast as a model as of the time of making this. However, that can change in just a few days. What I will say though is like my personal use case or my setup might not be ideal for a lot of people out there. A lot of people ask me like, you know, what's a great free model and this is free whereas Claude Fable 5 isn't. Of people say, well, you know, I want an open source model.

[00:49:25]Something like Quent is is potentially better for that. And additionally like you can't generate images and videos with claude either. So my default might be claude but then when I'm using it for use cases like images and videos I could use Quen.

[00:49:40]And also I don't use Quen sorry I don't use Fable 5 inside Hermes agent because you have to pay for the API and the API is quite expensive but I will happily use Quen 3.8 and it's really good. I mean, if we come back to that example where we said for/ learn and we plugged in the guide, you can see that it's pulled in the information from that page. So, it's got like a really good web search feature here. Then it also says like, do you want me to save this note into obsidian? Which brings me on to the next section, memory. So, if you go inside the studio here, it's going to have limited memory on what you do, what you work on, what your projects are. What I would recommend instead is that you use something like Obsidian, which is a memory system that works across all of the different AI agents and models that you use. So if you're using Claude, if you're using Hermes agent, if you're using, for example, Claude Code with GPT 5.6 or Kimmy or for example Quen 3.8, then how do you sync up the memory across all of them? Well, the simplest, most powerful system that I've ever used is Obsidian. And it's free, right? It's a free app. You can download it. It's open source as well.

[00:50:44]And then literally all you say to your main AI agent. So it could be Quen for example is you just say okay create an Obsidian vault locally with all my memories and everything you know about me. And you can ask for example chatb as well to give you all their memories that they have stored on you and then you can save them to obsidian too. Obsidian looks like this once you open it. So it's got the knowledge graph over here and every single one of these folders is basically just a markdown file. So if you click on that for example, you can see it says right this is a llama. Here's what it does. Here's how Julian uses it. Then everything links to each other, right?

[00:51:17]So it's kind of like a Wikipedia for my brain and it's all visualized inside one beautiful place. So the great thing about this is that we can always come back to everything else. We can have a look at everything. We can see what we've been working on. We can see how it's visualized. And then what I like to do is every single agent that I use, including for example Quent 3.8, Okay, we can link locally to this system. So, we have this full memory system here, this full vault and then everything is just neatly organized and it's free. It syncs perfectly. Agents love markdown files and folders and that's all this is. As beautiful as it looks, it's just a bunch of folders and project and files. That's it. That's it. And then all of your agents understand you, your history. So every time a new model comes along, let's say Fable 6 comes out tomorrow or for example GPT40 comes out in the future. It doesn't matter what model drops, it doesn't matter if Quen 10 comes out because the systems stay the same and the models change all the time. So this how I can stay up to date with AI. is how I can have a beautiful system like an HOS with all the Hermes workflows and for example the memory system and Quen all plugged in one place and I can just come back to it and it doesn't matter what happens next. You know if like Fable 5 gets taken down which it did a few weeks ago no problem at all. We can switch the models so easily inside this system and everything's just synced together with the memory system as well. Right? If some new crazy AI agent or some new CLI comes out tomorrow that we need to use comes out, no problem. We can instantly from day one have very personalized outputs. We don't need to train or anything. So you can plug straight into our obsidian vault which is stored locally and fully personalized to me.

[00:52:56]Right? And also you can get your AI agents to set up for you. The other thing that I like is like any of these different workflows here. So for example like the video agent or for example the music agent or the game studio whatever you're using Quen can operate all of them. So Quen is just a brain. Her Hermes is a hands. And these are all custom workflows set up with both of those. And then you can for example like research latest topics from your website. You can find new keywords and then I can create content directly to my website in one single click. I can create videos. I can create music. I can for example create infographics for notebook. Whatever I want to do. I can organize a whole team of AI agents with Hermes and Quen 3.8 plugged in. And then they can build for me and I can preview everything that I've built over here.

[00:53:40]That's the beautiful thing about having an agent operating system where everything just syncs together. And that's if you want to get my setup. I mean, I spend about three to four hours a day coding it. If you want to get my setup, you can get it from the profile.com link in the comments description. If you just want to build your own, you can. But just bear in mind, like it takes a lot of time to build this up and set it up yourself.

[00:53:58]Now, a huge benefit to using Quen 3.8 with Hermes Agent is that it can operate different programs that may be too complicated or difficult for most people to use. So, a classic example of this that's just dropped this month is that Blender MCP is available with Hermes Agent. What does that mean? That means you can do 3D modeling for stuff like this. So, for example, let's say you wanted to create a 3D model of a promo shoot or advertising a project or a product or something like that. Well, Blender MCP just dropped this month. And you can do this for loads of different apps. There's loads of different MCPs that you can connect directly with Hermes agent. So what this means essentially is that you can run a simple command like Hermes MCP install Blender or you can actually use our agent OS. So if you go to Hermes and then you go to the MCP section here, you can add MCP.

[00:54:47]So you just click on add custom over here or you can install any of the approved agents and MCPs like for example Gmail or Unreal Engine or NA10 or Linear or Blender. Right? So you can plug these all in and then what you can do from there is you can use Heromy's agent to operate these systems. Now, you might say, "Okay, but that's great, but why don't I just do it myself? Why don't I just click the buttons myself and go over to Blender or whatever?" Well, a lot of these tools that are absolutely amazing, like Unreal Engine or, for example, uh, Blender or any of these other MCPs you connect, even NA10, very time consuming to learn, very fiddly to mess around with, very technical to actually use. I mean, most people try and learn something like Blender to even to create something as basic as this.

[00:55:29]Most people try and learn it and then give up halfway through, which is pretty bad, especially if you're a busy business owner because number one, you don't want to be doing the work yourself. Number two, you don't want to be hiring like a full-time Blender expert for just uh maybe a project that you create once or twice a month. And then number three, there's no way out of all the things that I could learn as a business owner, I'm just going to sit down and learn Blender today, right?

[00:55:54]That's that's not at the top of the priority list. However, if I can just in a few clicks get Hermes MCP to do this for me, which did all of the examples you've seen on this page today, well, the great thing about that is that it can skip the dozens of hours of learning. It can get the results done for me and if it's not good enough, it just selfiterates and self-improves. So, the beautiful thing about this is things that were not possible for me before simply because of the resources, all the time or the energy involved, all the priorities I have, these things are not possible. But now I can automate them with an AI agent like Hermes. And this is not just for Blender. It's for all the stuff that I've shown you today. You can connect custom MCPS as well. So there's loads of cool MCPS. For example, like we were talking about Obsidian and the memory system. You can actually connect MCP Obsidian to Hermes agent and it can control, automate, and operate Obsidian for you, the memory system. And so pretty much every single external app that you want your AI agents to control, you can build into your Hermes agent. So you can add a custom MCP anytime and then it can do all this amazing stuff for you in the background without you learning it and you just sell it in plain English you want. And I think that's one of the biggest benefits of using Hermes agent. And 99% of people won't unlock this with Quen because they don't know how. So if you're watching this, you're way ahead of 99% of people.

[00:57:16]And that's just another advantage that you have.

[00:57:20]Now, also some people say, "Okay, what should you spend your time on or what should you automate first?" A lot of people come to me, they're overwhelmed, they're struggling, they learn all about these tools, but they don't know how to really implement them inside your business. What I would recommend for you is that you look at where you spend your time. You write it down for like you just track your time for a week or even a day and you look at, okay, where do you spend all your time? What tasks are you working on? And then how can you automate those? So for example, if you are creating a lot of videos, it might be time to get a video agent. If you are creating a lot of SEO content, it's time to get the SEO workflow set up, right?

[00:57:56]Whatever you are working on dayto-day that can be repeatable and can be automated, just start there. And then if you do that once a week, you know, it might only take an hour to set up, but you do it once a week and by the end of the year, you got 50 automations set up.

[00:58:10]And it's all based on stuff you actually do that you're wasting your time on before. Also, you might say, "Okay, well, how do you ask Quinn or how do you use these a AI agents to set up the automation?" So, you just go into Hermes like so. Make sure you have Quinn 3.8 plugged in here.

[00:58:26]And you can just say, "Okay, spend all my time on this. Help me automate that."

[00:58:30]And it'll actually walk you through the steps in terms of how to automate it.

[00:58:32]So, that's what you can do here. So, that's literally how we set up Hermes Oracle. I just went into Hermes. I was like, "Mate, set up a custom workflow where you pull in the latest AI automation trending news every day and then you allow me to publish social media content and SEO content based on that single trending topic in like one single click and it did it." And then you just test it, you iterate it, give it feedback, and you go back and forth like that. So that is basically it. That is a full 1hour free course on how to use Quen. We've talked about the Quen chat, how to use images, how to generate videos, how to generate coding projects, how to generate apps, the differences between Quen 3.7 and Quen 3.8, how to use coder as well to use the ID locally and code out and build locally, and also how to use Quest, plus how to set up Quen with a memory system so it's all fully personalized to you with Obsidian.

[00:59:20]And additionally, how to plug an agent operating system into Quen 3.

[00:59:27]So you have all your agents synced up in one place. You got custom workflows. You know where to start. And also for example, you could plug in Quent 3.8 into something like Claude Code and you can automate and build anything that you want. So thanks very much for watching.

[00:59:38]If you want to get all of my best trainings on this sort of stuff. We have loads of trainings on Quen 3.8, Kimmy K3, etc. You can get that inside the air profitable link in the comments description or just go to the air profit.com. Inside the community, you can ask questions every day and I personally answer them every day. Plus the whole community is there to help you too. There's always people online ready to help you whenever you need to. Inside the classroom, you get access to all of our best trainings so that you can learn and build anything you want. If you're a complete beginner, you can go from beginner to expert over here. If you want to get our 1hour courses that's over here and also inside our new daily updates, we have the full agent OS system with a full step-by-step guide video tutorial zip file with Quen 3.8 plugged in. You can also get full systems on how to, for example, run an agent operating system for free, how to use Kimik 3. We've got loads of great courses, even how to reduce the amount of tokens you use with whatever model you prefer. Inside the calendar, you can jump a weekly coach calls, get help and support your time. And inside the map, you can actually meet people in your local area who are building with AI agents like when and other stuff. So, feel free to get that link in the comments description. You might say, okay, how's everyone else's experience?

[01:00:42]Cuz it's great for you, Julian, but how's everyone else's experience with the agent OS and all these other systems? So you have 201 pages of testimonials, reviews, wins, and just great positivity from all the awesome people inside this community. So it's a great place to learn. It's a great place to share your journey. The vision is just to learn and grow and enjoy the whole journey. If you're watching this, you are super early, which is awesome.

[01:01:04]And so you might as well take advantage of it. And this is a great place to just meet other people, network, learn, grow, and also I look forward to connecting with you inside there as well. Thanks for watching. Cheers. Bye-bye.