ChatGPT o3 VS Claude 4 VS VS Gemini 2.5 Pro: Who Wins?

I just put chat cheap03 Claude 4 and Gemini 2. 5 Pro through the ultimate coding battle to see which AI deserves your money. And the results will completely reshape how you choose your AI tools. You'll discover the shocking moment one AI's entire platform crashed mid test while the others kept running. Why OpenAI's massive 80% price cut might be the smartest or dumbest move they've ever made. and the brutal sidebyside coding gauntlet. They're exposed. Which premium AI can't even handle basic tasks on the first try? O3 versus Clawude for Opus versus Gemini 2. 5 Pro. Who wins? Today, we're going to be testing them side by side and seeing which one performs the best in a coding gauntlet. Now, Chat GPD have actually just released a new update where the price of coding with 03 is 80% cheaper. So, we'll be testing out and seeing which one gives the best outputs and which one performs the best side by side. So, we have Opus in the middle, we have Gemini 2. 5 Pro on the right hand side, and we have Chatch GBT3 on the left hand side. We're going to get straight into this and just run it through some examples.

First thing that we're going to try here is we're going to say build a bouncing ball physics game. You can get all the prompts inside the AI success lab completely free link in the comments description. We'll test this out. So we're going to say build a bouncing ball physics game where each bounce produces a musical note and splash of light. Add bonuses for combo bonus bounces unlock new sound sets blah blah. Right? So we'll code each of these inside claw gemini and chat. What we're looking for here essentially is not just speed, but also which one performs the best in terms of outputs and which one gives us the best UI as well, the best experience. Right? So, they're all coding out directly here. It looks like Claude is leading the way with chat GPT 03 coming in second and then Gemini on the right hand side performing. By the

way, if you're wondering, okay, how much do it cost to code with each of these if you want to use the API whilst waiting for these to load? So Gemini is the cheapest to code with. You can actually get a free API. Claude, Sonnet. Let's compare these two. So we can actually do a comparison on Open Router in terms of which one performs the best. We open up the models here. We're not using Open IO3 Pro cuz it's not particularly good as an API. It's just going to take 5 10 15 minutes to generate each output. We're just using the normal 03 model. So let's find that. And then if we compare these versus claude and you can see a comparison of the prices right here, right? So for example, open 03 versus claude. Opus the costs for opus are a lot more expensive than the cost for 03. But let's see which one performs best. So if we go back to claude now, we have the example. Let's pull this open. See what we got here. And that seems to have come up with an error. So that's not actually working. seemed to work like first time around, but then it totally failed. Let's have a look at Gemini 2. 5 Pro. That has also come up with an error and let's have a look. Open three. We'll preview this. Oh, that's working perfectly first time round. All right, it works off that before it makes any crazy noises, but you get the point. So, essentially 03 is actually beaten clawed Opus 4 and Gemini 245 Pro using exactly the same prompts, right? And you can see we tried to fix the error before on Opus, didn't work. on Gemini, doesn't work. But 03 actually performs better and it's cheaper on the API versus 03 versus Claude Opus, right? So big difference there. And you can see actually 03 is not a bad model to code with at all. So let's try something else. Now what we're

going to do is we're going to take the next prompt. Take this. We'll start a new chat in each so it's got a different context. We'll say make build a surf racing game where you fly a rocket powered blog post past AI spam sites. We'll say make it feel like a flappy bird style game. Add crazy colors, too. All right. And then we're going to plug that into 03. We'll do the same inside Claude. And then Gemini as well and see what we get back. All right. Just bear in mind that 03 was the only one that managed to do the task first time round. The other two didn't manage that at all. And also when we tried to correct the errors, both of them didn't work. So let's wait for them to load. I've not actually coded that much with 03. But I did want to check this out since the new update came out today with the 80% cheaper costs. So you can see the announcement of the 80% price cut right here from OpenAI developers and the costs have gone down way cheaper than before. Look at OpenAI 03 this month versus last month, right? The prices are just so much cheaper. So let's see what we get back here. We've already got the output from 03 which is pretty wild as well. So let's preview this bad boy. Hopefully it continues up the momentum but it's not looking good. Let's see what we got. Oh, there we go. It's not bad. I like the colors, the UI, the design is super basic, but honestly, AI is not great at coming up with UI, but that's worked first time round, and it was the first to create it. Let's have a look at Gemini. Now, to be fair, this is looking nicer. I like the emoji style rocket that we've got. No sound effects. Let's see what we got here. All right. And then let's test out on Claude. I would say Claude Alpa is the best to be fair. Like that's way crazier and super addictive for weird reason. So if I'm going to have to call a winner here, I'm going to say that Opus won followed by Gemini and then Open Eye 03 didn't do a bad job to be fair. Like it did something decent. Brandon asks which Gemini version is this. So this is 2. 5 Pro preview the one inside gemini. com and Brandon says please tell me the code is better with chat GBT. It takes a while to get it to do what I need to do. Claude definitely does it better from what I've been experiencing but I use both together. Yeah. So it's mix so far as you can see today. Like round one chat chy 031 round two opus one. So it's mix so far but we'll see what we get back so far. This is definitely looking a lot nicer so far. So let's see what we get on the next test here. I'm going to take all of

these and create a new example and I'm going to go into check GPT just to get a little prompt idea for this. So we're going to say okay create a prompt for a dopamine inducing AI growth calculator single HTML crazy colors visuals confetti links to my funnels we'll see if chat GBT4 can just give us a good prompt for the next test. Here we go. Now, what we'll do is we'll go back to each of these and get the party started. For some reason, Claude doesn't want to start a new chat, so we'll just type it in manually. And in the meantime, we'll go to Gemini as well. Claude seems broken. Is Claude down? All right, we'll have to use the open router if you can see here Claude is not loading on the new chat. So, what I'm going to do is I'm going to go over to chat inside here. Open router. Then we're going to add a model and we'll type in called Opus and we'll just try that instead. So 03 has come back to us now. Hey, it's looking pretty good to be fair. Let's just check it works. Yeah, that's looking incredible, isn't it? It's never been so fun to calculate. Let's click these go to my funnels as well, as you can see. Pretty nice. Yeah, not bad at all. 03 is doing well there. Let's have a look at Gemini 2. 5 Pro. It's not bad. I wouldn't say it's as fun to use as the previous one. And also, the charts don't really show up here. Additionally, if we have a look here, you can see that you can't get back to the previous screen, which is not ideal. So, you know, you can't calculate the time. And also, it feels a little bit laggy when I'm scrolling up and down here. You can see the lag a little bit. So, it's not as good as 03's. Let's have a look. But we got Opus here doing its magic. Claude. AI is not opened at all, which is annoying. I'm just going to close that for now. It's broken. And then we can test this out on LiveWeave. And let's see what we got here. So this is the output from Opus. It's a little bit laggy. Like you see how it's lagging here and it's like struggling to scroll down and stuff like that. Like it feels really laggy. It looks nice. but it's just not quite nice to use when you actually scroll through the website. Like people are not going to enjoy that. So if I had to declare a winner here

I'm going to say 03. So 03 won the first round and it also won the third round, but the second round Claude Opus won. However, the main website of Claude is not even opening up. If we go to Cord now, h it's working again now, but it wasn't working before. Overall, I'm saying Chat GPT03 has won that. We did three tests. You can see the outputs here. The website didn't fail unlike Claude. It won two rounds out of three. It created really nice outputs. You can preview it in the canvas as well. And everything's looking really good there. So, thanks so much for watching. If you want to get all the prompts from today, feel free to get that inside the AI success lab. Link in the comments and description. If you want to get access to daily new free trainings as well, you can get that inside the AI success lab along with a community of 10,300 people and also all of these different courses on the left hand side like you can see right and all the video notes from today are inside here along with the details on the new 03 model pro that just came out and you can get that link in the comments description. Appreciate you watching. If you want to get an AI automation session that basically looks at your business, helps you figure out, okay, what's the best way to automate your business to save you a lot of time, then feel free to get that link in the comments description. You can book in a free AI automation session. We can jump on that call together, help you out, give you some tips, and that's essentially it. Thanks for watching.