NEW Gemini 3.0 ULTRA? Quasar Alpha is INSANE (FREE!)...

Quaza alpha is absolutely insane. So what you can see here is this is a new cloaked AI model that has just come out. It's completely free. It's got a million tokens context which is insane. Doesn't cost you any money and from what I've tested so far is super impressive. So it's a new stealth model that's just come out from open router. I'll show you

how to plug it into different tools and test out as well. You can also use the chat right here. So just go to openrooout. ai, I get your free account and then you can sign up and start using it inside the chat right here. And a lot of people are going crazy about this, right? So, you know, some people are saying that it's kind of like a really new mysterious model. A lot of people

have been benchmarking it and saying it's blazing fast. Plus, it performs on par with claw 3. 5 Sonic, which is a expensive model and normally you would have to pay for that. And this is completely free, right? Quazar Alpha is completely free. I've also test out for coding is super interesting and open rout said excited to announce our first ever stealth model. All right, so it's a pre-release of an upcoming long context foundation model from one of the model labs. Some people saying it's Gemini 2. 5 Ultra. Other people are saying it is the new release from Open AI. Nobody's quite sure, but it is very interesting and you can see some of the stuff people have been creating right here. It can even create music I think some people are saying. Right. So you can see here what people are building directly. This is an upcoming model from OpenAI because OpenAI actually just said they've revised their road map and they'll be releasing 034 mini and GPT 5 within the next few months. And you can see what sort of stuff it can build, right? So, for example, it created this really powerful hexagon with vibrant bouncing balls. Seems to be pretty strong at coding. We'll test out inside this video right here. Matthew Berman, absolute YouTube legend, was posting about this as well. And he was saying that it's a 1 million token context model beating claw 3. 7 Sonic on benchmarks whilst running four times faster. How crazy is that? 4x faster. All right. And you can get it available at open router. you can get early access. I also show you some of the benchmarks here so you can see how it's performing and it's performing really right now. And additionally, it is completely free. Okay, so if you want to get access to it, you would just go to open routter. ai. Then you're going to go to the chat and just select quazar alpha. All right. Now, if you can't find it, just type it in the search model section here. And you can see comes with 1 million context tokens. super fast. Loads people using it within root code and client which I'll show you how to insert and we'll test it out in a second as well. I've been blown away by how fast it is. If we go to the chat right here, we can do some examples. So, if I

take an example prompt, we'll say okay, generate a HTML file with vibrant bouncing balls in a hexagon. Wait for that to load. You can see how fast it is, right? So this is coding at the same level as something like claw 3. 7 sonet or 3. 5 sonnet according to the people that I showed you on Twitter a second ago. And if we copy this and test it out, we'll plug this into live weave and just see how it performs. All right, so you can see an example right there. And you can just build out whatever you want using this process. All right, let's take another one. So, we're going to say here inside a new room, new chat using Qua Alpha, create an inter interactive water molecule simulation that visualizes the formation and breaking of hydrogen bonds in real time. You can see how fast it is absolutely rapid. All right, so it's really fast and you can also use the API from this as well, which I'll show you in a second. So, you don't just have to use the chat. Let's grab the HTML file over here. Plug that into LiveWeave again. And look at that. It's just working perfectly, right? And then we can speed up the temperature. There's no lagginess or anything like that. I mean, if we compare this side by

side with something like Claude, so we'll add a model here. And we'll select Claude and we'll just see how it performs, right? So, we're going to use exactly the same prompt. We've got Claw 3. 7 Sonet and we've got Qua Alpha. And let's see which one is faster and which one builds a better output. So, you can see how fast Qua Alpha is versus Claude, right? This seems to be a lot quicker. Like this is pretty much finished the coding whereas Claude is taking his sweet time. Is that actually is Claude even coding anything? There says I've created an interactive water simulation but there's nothing here. If we open up the example from Quazar actually went off and did its magic. All right. So if we compare it side by side, Claude 3. 7 didn't even create anything whereas Quazer actually did. Let's test it out directly inside Claw 3. 7 Sonet inside the chat feature. Maybe we'll get better outputs here. So, I'm going to delete that and then we'll grab the prompt here. Probably cuz I didn't type the prompt in properly and we'll see which one performs the best. All right, but you can see that claw 3. 7 set is a lot slower. And also, if you want to code with claw 3. 7 set as an API is going to cost you money each time you use it. Whereas you can use Qua Alpha for free directly inside the chat here. So it's taking it sweet time. Yeah. So you can see this is a lot faster using Qua and the outputs are pretty good as you can see. Let's test one more now. So

I'm going to create a game with Quazar. We'll say okay make me a captivating endless runner game. Key instructions on the screen. P5JS scene no HTML etc. I like pixelated dinosaurs and interesting backgrounds. To be fair, I think I still prefer the output from Claude. You can see here just like the UI is a lot nicer, but if you're looking for a free API to code with, then Quazar could be better, right? And then we've got the P5JS ready to go. Let's test it out. If you want to test this stuff out, just go to editor. p5js. org. Plug that in. We can hit play and we have that ready to go. Right, that seems to be working. There we go. Look at that. Codes blazing fast. Creates games, creates tools, simulations, etc. And we're living the dream. I don't I still don't think that's as good as something like Claw 3. 7 Sonic or Gemini 2. 5 Pro from what I can see, but I love the speed of it. Like the speed of it is absolutely insane. All right, so if you want to

start coding with this as well, what you can do is you can go over to keys over here, create key, and then just create a key for OpenAI. Once you've done that, then you can download something like Visual Studio Code. So go to code. visisualstudio. com. If you're struggling to set anything this up, join the AI profit board because it shows you exactly how to set this stuff up. Then we're going to open up Visual Studio Code directly here. We'll go into something like recode settings. And then inside settings, we want to select open router. And then inside open router, we're going to select alpha. And we should be able to code with it. You can see open router quazar alpha right there. All right, we'll hit save. We're done. Boom. So, just note here as well. So, it supports images, but it doesn't support computer use or support prompt caching. All right. And then let's try for example prompt right here from the AI profit boardroom. Let's just we'll create something basic right now. I'll create something like generate then SEO cost calculator website simple one file HTML just to keep it simple and easy so we can show you how it works and you can see that works really quickly right here. Now we're using boomerang mode as well. I've got a video on that yesterday if you want to check out how I use Boomerang mode to code. But inside Root Code we can just build out and code whenever we want using this process and we can do this all for free. Visual Studio Code is free. Root code is free and then the API from Quaza alpha is free as well and then you can just build whatever you want but it'll be very interesting to see is this a model from for example the new Google updates or from OpenAI I don't know but the internet is going crazy over it. So it's now coding that out and this is the great thing as well about AI now is you can build and code even if you don't know how to code and you just have to give it an idea. So, if we hit run command over here, this should open it up. And we've got the SEO cost calculator ready to go and it's built out right there. All right, let's test something else out now. So, we're

going to say, okay, create a self playing snake game. It now splits everything into subtasks. Again, I would definitely recommend checking out my video on Boomerang if you want to learn how to build out stuff like this like almost seamlessly. So, you've got a full sort of army of AI agents building out subtasks together. But you can see how this works and how quick it is and you can also use open router and quazar inside client as well. So, if we go over here and you select open router quazar alpha, you can get that set up. Then we're going to hit save and let's see if this runs properly. All right, so it's created it ready to go. We'll run the command. That one didn't work so well. So, I do think it's not perfect, right? So, if we check this out, like you can see how you can't actually play. I'm actually going to say this just says game over and is playing. So, let's see if it can fix itself as well. Now, it's creating more subtasks with the boomerang feature. This is the biggest difference as well. If you're thinking about coding with R code or with client, root code is really good because it uses that boomerang feature to segment everything into separate tasks. We've hit the quot limit on that. So it seems pretty smooth up to that point. That is not ideal. Let me test out if we can do it inside the chat right here. We'll see how good it is. All right. So if we go to open router quazar chat and we'll just say now okay create snake game in HTML and see if it works better inside the chat cuz it's hit the quotota directly inside the API. All right. So, it's already come back to us with that. Let's make sure this actually works. Plug that in. Yeah, that works perfectly. All right. So, if you ever run out of the API credits directly inside root code or client, then just go back to the chat inside open router and it seems to be a lot more smoother. Actually seems to give better outputs inside the chat than inside code. Code didn't work the first time we tried this, which is quite interesting as well. But yeah, there you go. And then this will keep retrying until you actually get back the quotota. Right? So you might have to wait a few seconds. And now that it's working perfectly great. All right. So you know how to self-correct stuff. How to build games, apps, tools. You can use this for landing pages, websites, etc. Or you can just use it as kind of like a free alternative to claw 3. 5 sonnet or 3. 7 Sonnet. The limits don't seem to get hit inside the chat as much. Whereas if you're coding with the API, that seems to use a lot more credits. you seems to have a limit on the quotota. So, thanks

very much for watching. If you want to get all my best prompts, tips, workflows, etc. for using APIs and setting up open router or client or for example root code as well, feel free to get that inside the AI profit boardroom. This is a community focused on making more money and saving time with AI. So, if you want to scale your business, if you feel like you're behind, if you want my best sort of click and deploy AI agents and systems, you can get that inside the AI profit boardroom. For example, like if you just want to set up an AI powered short form generator, you can download the workflow here and then just upload it straight into NAT. All of these modules right here have different templates, workflows, etc. Right? There's even a crash course to say punches vows of AI. And you can see this gets updated daily with all of my best systems, prompts, and workflows. And on top of that, this comes with weekly Q& A calls. So you can jump on the live Q& A, ask any questions you have. And if you get stuck in between the Q& A calls, you can ask questions inside the community and get help directly right there. All right, so feel free to get that link in the comments and description. Prices are going up this month, so make sure you lock in your legacy price right now before you miss out. And you can see the schedule for weekly calls right here. Additionally, if you want to get a free one-to-one SEO strategy session that shows you how to get more leads, traffic, and sales with SEO, feel free to get that link in the comments description. We'll basically show you how to get more free customers, scale your business, and make more money from Google. All right, so this shows you how to go from zero to 145,000 business month based on what's working for our clients and for us, as you can see from all the testimonials right here. And also on that call, normally you're paying SEO consultants like thousands of dollars per hour. Now, you can book in a free one-to-one SEO strategy session and it's completely free, right? Just jump on the call. You got nothing to lose. So feel free to get that link in the comments description. Appreciate you watching.