Fusion DESTROYS Fable 5?

Key Takeaways

So, we have a new model out from OpenRouter. It's an API that called Fusion that is designed to achieve favorable five-level intelligence. And today I'm going to show you what I've built with it. Here's one example, and it is blowing me away in terms of how cool this is. Bear in mind this was created with one single prompt. All of the stuff that I'm about to show you is created with one single prompt. Let me show you another example here. So, this is pretty cool. This is fun to play. It's kind of like a a 3D game where you go through a maze. And then let's have a look at this one. This is a dragon realm, like you can see. Pretty cool, fun to play, etc. Full open-world game. This is kind of like a basic, you know, break the blocks game, but it seems quite fun to play as well. And I'll compare it against all the other stuff. So, we've run the same prompt on on many other models, and I'll show you the difference in a second. And this one is really cool as well. This is kind of like a a neon racer game, as you can see. Super futuristic, really interesting to play, and a lot of fun. So, let's run through what Fusion is first of all and how it works, and then we'll come on to how it performs versus other tests and where it scores on the Goldy bench benchmarks. So, let's get into it. So, this is Fusion. If you're wondering how Fusion works, essentially what you have is five models that work as a panel, and then their answers to the same prompt go through a judge who fuses the answers, critiques them, is kind of adversarial with them, and then gives you one single answer. So, the way this works is really it's one shot because you use the API once, that comes up with five answers that the judge then turns into one answer, and then everything that you saw a second ago was created with one single prompt, which is pretty crazy in itself. Now, in terms of the benchmarks, how does it actually perform? By the way, just want to say here, I've got Goldy bench and I've shown you the benchmarks, and I'll go deeper into those in a second. These are benchmarks from Open Router. Test it yourself. See what you think. Don't Don't just listen to benchmarks. And you can see the announcement from Open Router here. So, obviously, Fusion came out just after Fable 5 got taken down. And on their tests, basically, when you're running multiple models together, the idea is multiple minds are better than one mind. So, when they work together like this, you get better answers. And you can see how they perform. So, it's designed to achieve Fable 5 level intelligence or higher. Now, if we look at the tests from Open Router, you can see how they perform. So, solo Fable 5 does pretty good, but GPT 5.5 with Opus 4.5 synthesized, aka you've got the judge model with Opus 4.8 looking at the answers and fusing them together, that tends to work the best. Now, obviously, Fable 5 and GPT 5.5 synthesized by Opus 4.8, that is goated. But, obviously, you can't get access to Fable 5 right now, so it's sad times. But, you get the point. Now, the other thing is, well, that if you are running an API, it can achieve Fable 5 level intelligence on benchmarks whilst being 50% of the cost. So, when you think about that, it's pretty wild. And basically, one API fuses the best output of multiple models. That's how this works. Now, in terms of the benchmarks, we've tested 42 tasks with Fusion so far. Draco Deep Research is score 69 with a budget panel. So, bear in mind that you can have a budget panel or you could have a premium panel. And what I mean by this, like, you could have cheap models running side by side, or you could have expensive models running side by side. And obviously, if you use the expensive models, you're probably going to get better outputs. But, either way, it should be Fable 5. Now, you wouldn't use this for like everything, right? If you're using Fusion in a panel of APIs and a panel of models working together, you're You just going to use this on the big tasks. This wouldn't be like a day-to-day API that you run. But it's built some pretty awesome stuff. Here's another example. So, this is like a pool game emulator that actually looks really good. And we created like a voxel dash game. As you can see right here. And again, like the outputs are super nice. Just It's fun. A lot of this stuff that it's creating is like fun, looks really nice. We've used it on websites and stuff like that as well, but you know, obviously, I like to show the visual stuff that's fun to play with. Now, how does it perform versus everything else? So, if you have a look here, for example, we've got Fusion versus Opus 4.8. And we can see side by side how they performed in terms of the actual outputs. So, this one is Fusion. And this one is Opus 4.8. So, let's have a look at Opus 4.8 first of all. It's pretty basic. Like, it does a basic sort of job. Not that interesting to play cuz the ball is like behind the wall half the time. But you get the point. That's how it works. Now, let's have a look at Fusion. So, you can see it running right here. A lot more interesting. A little bit laggy on the gameplay. But a much more interesting game. And you can see the moving background as well. Next up, we have the ray cast game. So, here's an example of the one from Fusion, as you can see right here. Pretty cool to play. Works really smoothly. Again, this was just created in one prompt. Which is unbelievable when you think about that. Like, we didn't have to go back and forth. We didn't have to test it. It just works straight out of the box. Now, let's have a look at the one from Opus 4.8. It's a little bit like buggy. So, you see how we can kind of run into this wall in the graphics are a bit weird. It still looks good. Don't get me wrong. It's a little bit minimal as well. Here's another one. So, this is Out Run. Again, this is the one from Fusion. Looks really cool. Cool outputs. A lot of fun to play with. I will say though, when you're using for example Claude, the problem with Claude, sorry, the benefit of Claude is that you can use that subscription with your CLI. When you're using these API models, obviously you're limited to the API. You can't use it as a CLI, for example. Now, this is the output from Claude Claude Opus 4.8, which is kind of sideways the whole time, so you can't control it properly. It's still fun to play, but what is going on here, you know? Bear in mind that you can have Opus 4.8 working inside Fusion, either as the synthesizer or as the actual model itself. Now, one final thing I want to show you is that what we've actually done here is we've plugged it in the agent operating system. I think this is one of the best ways to use it. And so, what you can actually do is you can create something with Claude, plug that into your workspace, and then you could plug that into Fusion to improve it or to compare it side by side. So, you can see, for example, we can chat and use Fusion over here takes a few minutes to generate a response. We can ask questions here. And then, once we've generated that, we actually get the outputs inside our workspace, as you can see. So, anything that we create gets saved inside the workspace, and then we can come back to it later. But, out of everything that I've seen, it creates really smooth, cool stuff. I mean, it makes sense, right? Like, if you can have the best of every single model, and if you can avoid the mistakes that each model makes by comparing them, like obviously you're going to get better outputs. So, it's really cool, it's fun to play with. Would I say it's Fusion-level, sorry, Fable 5-level? I think it's definitely up there. Definitely up there. But, I think also it's worth you testing it out yourself. And also, bear in mind, like, if you're using Fable 5, obviously you would have the Claude CLI. And so, you could go back and forth with it a lot more without having to wait for output. So, one of the biggest disadvantages is like, it takes so much time to use Fusion. That is not ideal. Like to ask a question, come back in 5 minutes. It's great to have amazing answers. It's great to have amazing outputs. But, it's not that easy to use. There's a lot of friction to this system. So, thanks so much for watching. If you want to get all of this stuff, you can get it inside the AI Profit Boardroom. We'll keep testing models, sharing what we've created. We have the benchmarks right here. And I will say Fusion is scoring the best out of everything on all of the tests that we made. I will say as well, GLM 5.2 really good if you haven't checked it out right now. Definitely worth having a look. So, thanks for watching. Feel free to get the AI Profit Boardroom and the Agent Operating System inside our community. You can ask questions and each day I answer them with video tutorials personally and also the whole community helps each other, too. So, there's always people online 24/7 so you can get help and support whenever you need it. Inside the classroom, you can get access to all my best training. So, if you're a complete beginner with this stuff, you can just start with the beginner to expert course here. And if you want the Agent OS, it's right here. We've actually just added a new version today with the zip file and a full guide and video tutorial on how to install it. We have new daily updates and tutorials with guides including a full deep dive on Fusion here, as you can see. And as new stuff comes out, we share it actively. Inside the calendar, you can join our weekly coaching calls, get help and support in real time, share your screen, meet other people building with similar things. And then inside the map, you can meet people in your local area who using AI agents like you. So, hope to see you inside there. Link in the comments description or go to the AI Profit Boardroom.com. Fusion really good, definitely worth checking out. Tops the leaderboards from my test so far. I'll see you on the next one. Cheers. Bye-bye.