Build anything with DeepSeek-R1 0528, here you go!

Deepseek R1 0528 how to build anything. Today I'm going to show you multiple different ways that you can use Deepseek R1 to build basically whatever you want, including how to do this all for free.

So one of the best ways to get access to Deep Seek R1 0528, which is the latest update, just came out a few hours ago, is directly inside Open Router. Now, if you want to just build out something basic, for example, like we built out this dopamine drums game, you can get that inside the chat. So, you can just give it a prompt like you can see, then you can copy that and start using it pretty much instantly. So, for example, if we plug this in, you can run this via liveweave. com if you have the raw URL, but you can basically just start using whatever drum machine you just built directly here. Here's an example. So we said build a beat racer game where you control a ship dodging obstacles to the beat music and then you can copy that HTML directly here. This was another game that I built out with flashing keywords. You can start the game really easy to build out and yeah you don't have to build games and stuff like that. You can build apps or tools or whatever you want. But basically inside here just make sure that you have Deepseek selected. If you want to use the free version just select DeepSeek R110528 free and then you can start building with this. So that's the easiest and fastest way to start using it inside the chat. As far as I'm aware, it's not available on deepseat. com. There's been no official announcements of the latest update coming to deepse. com. So I think this is one of the easiest ways to get access. Now what you can also do inside open router and this will allow you to build a whole lot more. So if you go to the top right and then go to keys, you can get an API key and then start coding with this. So, for

example, if we go over to Visual Studio Code, inside Visual Studio Code, we can start coding. If we go down to Root Code, like you can see, then we're going to go to settings. Make sure you plug in the API key that you just grabbed from Open Router. And then inside the drop down here, we can select 0528. And you can choose between free and normal, right? So, standard, you're going to pay, but it's super cheap anyway. Or if you want to use a free version, you can select DeepSk R10528 colon free in the dropown. Hit done. And now you can start building whatever you want. So for example, a super simple example, but if we say build me a flappy bird game in HTML, hit enter. We can use R code and a combination of Deep Seek 10528 to start building out for us. Now, let's see what we get back and see how powerful it is. From the tests that I've done so far, Claude for Opus is still the goat, right? I don't think Deep Sea Gar One even compares in that way. But when I've tested it, and you'll see this in another video, Gemini 2. 5 Pro actually performed worse than Deep Seek R1 in sideby-side test. So, I've got another video showing you live exactly how that works, but essentially this is a really powerful model and it's very good at just getting one shot outputs. So, that's one way. Now, whilst we're waiting for that to load, another way that you can start using this, let's just test this out. I want

to see if it's available so you can build agents inside NA10. So, if we go inside NA10, we'll log in here. And if we set up a new chat trigger to automate an AI agent and I've got loads of training inside the AI success lab link in the comments description if you want to learn about NA10. But if we type in AI agent like then from here we're going to click on the plus sign. We'll select open router as the chat model and then we can select the latest update from deepcar1. As you can see, it's a 0528. Inside the drop down here, we'll just add a system message to make sure this works. And then we'll go inside the chat and just say, are you working? I might need to plug in an API key. Let's double check. Yep. So, we need to add it and upgrade. And then that's how you can easily build out AI agents directly inside the chat here. So, if we say, okay, are you working? There we go. So it's running the AI agent with Deepcar1. Obviously, if you're using Decar 1, it's a reason model, right? So you want to use it more for like coding tasks and that sort of thing, but you could easily create a coding agent inside NA10 to build out whatever you want. And then you can see using the Deepseek R10528 method, we've got the response in the chat right here. So it's yes, I'm up and running. I'm ready and available. Blah blah. And then also if you want to use the free version, just make sure you select the 0528 colon free model right here if you want to build stuff out. But that's how easy it is and that's how it works. So

just to recap here, we've covered how you can use this inside open routters chat for free. We've shown you how to build out with Visual Studio Code for free. And if you want the full recipe on setting up Visual Studio Code, you can get that link in the comments description to the AI success lab. And then finally, we've also shown you how to build AI agents using Deepseek 1 free with the 05283 model. Now, let's open up the outputs from Visual Studio Code. So, just to recap here, we said build a Flappy Bird game in HTML. Went off, build out. We'll hit save on this and then we should be good to go on that. Let's see what we get back. So, it actually run the command for us as well directly here. Let's run that. And we have the Flappy Bird game ready to go. Pretty hard to play to be honest with you. I'm not very good at this, but there we go. All right. Yeah, there we go. Boom. So, AI agents built, Visual Studio Code ready to go, chat ready to go. What else? And

what we could also do is if we set up something called BoltDIY, we can start coding with that as well. Right? So I've actually got a full SP inside the AI success lab where you can get access to this stuff. As you can see, it shows you multiple different ways to run DeepCar1. You can get the SP right here. Basically, if we go inside the SAP and we go over to let's have a look here what we got. We got Bolt DIY. So, we'll go to the SP for Bold DIY and then we should be able to start running this in our terminal. I'm just going to close Visual Studio Code so it doesn't take up too much memory. There we go. And now we can start building and setting up BoltDIY and from there start coding. So, if we copy this from the instructions, open up terminal, grab this. Basically, what BoltDIY is, it's kind of like a open- source version of Bolt. where you can build and code like websites, apps, tools, etc. So, we're going to start setting this up. You just have to run it in terminal first before you can start using it. But it's a simple case of just copying and pasting. Pretty simple and easy if you're using the SAP from us. And then we'll run that pmpp rundev. That's going to give us a local URL where we can host it like you can see. And then if we go over to here, we've got Bolt DIY ready to go. We just might need to plug in an API key, but let's test this out first. So if we go down to the DeepSsee list, we've got 0528 over here. We can use a free version if we want to. And then you can build a simple app. So just is a very simple app. Let's say build a to-do list app. We might have to plug in API key. Let's see what we get back. So just to clarify here, we're using like a free version of bolt new that's running locally. We've connected deepsear1 0528 for free inside the chat and it might take a little while for it to come back to you simply because it's a reasoning model. Right now if you changed that API down to like Gemini or Google or that sort of thing, then it would be much faster. But I want to show you how to use deepse because that's why he clicks on the video. So let's wait for that to build out. So now what you can see is it's actually coding out our project files and the good thing about this is all your chats are saved locally like you can see on the left hand side here. So all your old chats see you know all the old projects etc that you built with bolt DY are available inside the chat history. Now if you want to run this stuff locally from what I saw Deep Seek R1 the latest version was just updated an hour ago right? So you can actually use the latest version of Deep Seek R1 and run it with O Lama. We'll come on to that in a minute and how you can do it. It is quite a big model, not going to lie to you, but it was just updated an hour ago, so it's brand new. We'll come on to that in a second. And then inside bolt new, this is how you can code stuff out directly. And we'll just wait for that to load. Now, if it doesn't work or you get an error, etc., You can just go back and forth in the chat over here. And also if you want to see, okay, how well does this perform versus everything else, right? So this is a interesting tweet by Vibbath who says, "The new Deepseek R10528 is competing with 03 high and 04 Mini on live codebench. " So you can see the benchmarks here. So Deepseek R10528 is coming in fourth behind 04 Mini, 03 High, and 04 Mini. But it's actually ahead of other models, for example, like Gemini 2. 5 Flash. So let's check this out on live code bench to see how it performs. Yes, you can see on the benchmarks it is performing really well on the live code bench. So basically this is a holistic evaluation of large language models and you can see Deepseek R1 is coming in fourth over here and that is above just to give you a breakdown here. It's above claw 3. 7 sonet, above claw 3. 5 sonnet, above gro 3 mini, and it's placing above quent 3 as well, right? So, it's right at the top of the leaderboard here. If you look, deep sea v3 is all the way down, and deepcar10528 is absolutely crushing it. You can also get access via huggingface. co. If you go inside the inference provider section here and then just type in whatever you want in the sentence, you can start using it directly there.

And then if you want to start using this locally, install this locally, pretty simple and easy to do. So what you can do is you can download O Lama which is completely free. I have a full SAP inside the AI success lab on exactly how to set up etc. But basically you can download O Lama like you can see and then you can run Deep Seek 1 and other models like Quen 3, Llama 3. 3, Gemma 3 etc locally. And to do that, once you've installed O Lama, just make sure you have it running. So, let's open this up like you can see. So, we got Oama 2 running in the background. And you'll see it right here. And then once you've got it running, we can open up the terminal. And if we type in the command, for example, like run DeepSseek, which you can copy from here, you can run that in the terminal to install the model locally. So, you can see it's installing right here. And it will take a little while for that to load. But basically, we are now running this locally using O Lama Deepcar1. Right. So, you can see this was just updated an hour ago with the latest model. It's completely free to run. We're running it via O Lama. And all you need to do once you've got alarm installed and running is copy that go into terminal paste that command into terminal and then you can see it is now running perfectly. So if we say for example okay what is 2 + 2? It's actually pretty fast when it replies but you can see it now comes back to us pretty much instantly. Right? So that's another way that you can run and build anything with it which is directly inside the terminal. So we've

talked about five different ways you can use deepse one. Number one inside the chat for basic stuff like writing coding etc. You can also use visual studio code with root code. From there you can use bolty. You can also use o lama and you can grab the API key and plug it from open router into m10 and then build your own AI agents as well. If you want to learn how to use all of this stuff step by step, I have a full training course on exactly how to build anything with Deep Seek R1 plus eight different methods of using Deep Seek R1 so that you can build and code of anything. Um, today we specifically focus on the latest update from Deepseek, which is 052A. So, if you want to get all my best trainings, SPs, etc., feel free to get that inside the AI profit boardroom. This community focused on making you more money and saving you time with AI. comes with an awesome community where you can post and ask any questions you've got. We also make weekly automations based on what people want to hear and want us to build. So, for example, yesterday, so for example, we got 25 comments on this post inside the AR profit boardroom where people were like posting, I want to see this, that, blah blah. And for every single automation, we actually created a full workflow plus a video tutorial on exactly what to do and how to build it out, right? So, if you want us to build you an automation or help you on any sort of automations, feel free to get the AR profit boardroom. Massive value right there. And also, you get all our best courses, coaching, test templates, AI agents, etc. And this community is really just focused on one thing, which is making you more money and saving you time with AI. So, feel free to get that link in the comments description. Additionally, if you want to get a free AI strategy session that shows you exactly how you can automate the most important tasks in your business with AI, plus how to actually implement that. Feel free to get a free AI strategy session link in the comments description. We actually do have quite a lot of AI automation clients right now. So, if you want to become a client of ours, feel free to book in a free AI automation session. Link in the comments description and we'll be happy to help you. Thanks so much for watching. Appreciate it. Better.