Hermes Mixture of Agents DESTROYS Claude Fable 5?

So Fable 5 versus Hermes Mixture of Agents versus OpenRouter's Fusion, who wins? Today, we've tested them out and we'll show you the results in a second. This is something we actually built with Fable 5 a few minutes ago. So let's get straight into this.

We've compared them side by side and I'll show you exactly what the difference is. Now, if you're wondering, OK, what is Hermes Mixture of Agents? This is a way of combining multiple models together to get better outputs. So you normally get with, for example, something that's Frontier.

So, for example, if you wanted to create outputs that are better than Opus 4.8, you could combine GPT 5.5 and Opus 4.8 and that tends to work better. And that's something that I've tested personally. It's also on the benchmarks too. Now, Fusion works in a similar way.

So with Fusion, you basically have multiple agents working together as a panel. As you can see right here, then a judge fuses all the answers and they go into one answer. And of course, Anthropx Fable 5 just came back again. And we've tested it relentlessly with 42 different builds, which I'll break down in a second.

And you can see how they perform side by side. So let's get straight into this. One of the first things we did was test out a game side by side. So let's have a look at this.

So this is the example from Hermes Mixture of Agents. This is from Fable 5 and this is from Fusion. Now, out of all the tests, we have a look side by side. I would say that Fable 5 actually created the best output so far.

Bear in mind, the other tests with Fusion and Hermes Mixture of Agents were using non-Fable 5 models. So, for example, GPT 5.5 with Opus 4.8. And it's created the nicest output, as you can see right here. Next up, we have the Nordic Crypt game.

So this is from Fusion. Super dark, pretty hard to see what's going on. It kind of works, but doesn't work at the same time. Like, you wouldn't play that game at all.

If we have a look at the output from Hermes Mixture of Agents, quite nice, works nicely. A little bit hard to control, but it does work well. I also like the moving background as well. It looks quite nice.

Good vibe to it. Now, if we have a look at the one from Fable 5, I would say this is the number one, the easiest to navigate. Number two, it's got a lot more to it than the other two. Like, the other two seem to be a bit more basic.

This has more detail as well, and it's easier to navigate and move around. So, for the Crypt game as well, I would say that Fable 5 is winning right here. So, the interesting thing about this is, if you look at Hermes Mixture of Agents, if you look at particularly Fusion, which Fusion claims to beat Fable 5 on benchmarks, that's not what I'm seeing so far. But there's many more tests.

to try out. Then we've got the orbit test from our agents too. So, this is with Fusion. Not bad at all.

It's in a solar system orbit. So, it's like a diagram of the planets and everything else. You can change the speed and the controls over here. Then we've got the version from Hermes Mixture of Agents.

And you can also drag and move items into the galaxy as well, which is pretty nice. You can change the speed over here. The controls are good. You can reset it.

You can pause it as well. Some good controls on that. Now, if we have a look at Fable 5's output, I would say this is the most basic. So, it's not always the case that Fable 5 wins on every single test.

As you can see on this one, it didn't do so well. I would say Hermes Mixture of Agents won that. So, we have three tests in. Fable 5 has won two of them.

Hermes has won one of them. Fusion hasn't won any yet. Next up, we have the landing page test. So, this is with Fable 5.

And this is actually something I've noticed with Fable 5, is like it kind of struggles with UI. So, if you look at this particular output, it doesn't look so nice. It's a little bit messy. The background is moving, but kind of feels weird.

Everything's just a bit off in this particular example. Now, if we have a look at Nova, this is the output from Fusion. It's definitely beating Fable 5. So, it definitely outperforms Fable 5 on landing page creation for this particular test.

Now, if we have a look at Hermes Mixture of Agents, I would say this is by far the nicest output that we got. So, you see how it's glowing? When I move the mouse, the background is moving, but it still looks quite nice. We can click it and it's got these cool animations as well.

And by far, that is the website that I would prefer to use out of all of them. It's still a little bit messy at the bottom, like this is not properly aligned. So, all three of them were not like the best in the world at creating websites. But if I had a choice, I would go with Hermes Mixture of Agents.

Now, at this point, you might be thinking, okay, why is Hermes Mixture of Agents holding its own against Fable 5 when Hermes is using older models and Fable 5 is a newer frontier model? Well, I think this is because every single draft gets a second opinion. You know, when you've got a panel of minds and they're working together, basically you've got one judge that fuses the answers and quality controls it all. Then you've got a chair that works together and basically looks at, okay, what's the best of each?

Let's take the best idea. Let's take the best options from each piece of code and then plug them all together. And also, when you've got a panel, so when you have that judge that's looking at the other agents' work and the other models' work, basically it's rooted by evidence, right? So, it will kind of like critique the outputs of the other agents to get better outputs.

But let's keep going to see where they perform on everything else. So, another one that we created was Dragon Realm. So, this is the version from Fusion. Pretty nice, looks good.

The Dragon's a little bit off in terms of the graphics, but overall it looks really nice. Even like, for example, the snow that's falling, like, looks pretty good when you're navigating around. I will say it's a little bit difficult to navigate around. It's like not that smooth, but it's pretty good overall.

Then if we have a look, this is Hermes Mixture of Agents, which is pretty good, very similar to the other one, but I would say slightly better from Hermes Mixture of Agents. Just because it feels more open, it's easier to navigate, but they're pretty much on par. Then let's have a look at Fable 5. So, this is the output from Fable 5.

And again, very similar. The only difference I would say between them is that this has a nicer intro where it just moves around without you. But overall, I would say that's a draw between all of them. All very similar.

Next up, we have a test with a web OS. So, this is an operating system we created where basically you have all of your applications plugged into one system. So, this is the one from Fable 5. This is from Hermes, and this is from Fusion.

So, let's try this out. We've got a notes app here. Let's test out typing something in, closing it, and then seeing if it auto-saves. So, you can see it doesn't save on the Fusion version.

So, it kind of looks nice on the outside, but then when you actually test it, it's not saving anything, right? We actually can't click the file to save section here as well. So, it's not auto-saving. If we try and close that, it doesn't seem to be that responsive.

The paint section seems to work well. Terminal works okay. And we've got the calculator as well. But we can't really use the equals button.

Like, pretty hard to navigate. Like, you can't just scroll down, if that makes sense. So, not bad, but not great. Then we've got the version from Hermes mixture of agents now too.

This looks pretty good. Let's just test the notes and see if this actually saves. So, it's being a little bit buggy. You can see the notes application is moving around the screen, but I'm not asking it to move.

Paint seems to work okay. The background is super nice. I think the UI is actually nicer for Hermes mixture of agents, but it's not very responsive. It seems very buggy, right?

So, if we click on this, for example, or if we try and move the notes, we actually can't. It's kind of broken. Let's have a look at Fable 5 now. So, Fable 5 seems to work pretty nicely.

So, look at the paint section here. Yeah, it looks good. So, look at the terminal. Can we move that around properly?

Yes, we can. Yeah, all of it is working. It's not buggy at all. So, I would say in terms of that test, Hermes mixture of agents came in last.

Fusion came in second, but Fable 5 is by far the best. So, if we look at the results so far, we have Fable 5 winning on three different tests, Hermes winning on two tests, and Fusion is not winning on any. Now, we're going to try something called Voxel. So, this is a version from Hermes mixture of agents, and it's super buggy.

I'm actually going to close that because I literally can't navigate, and it keeps refreshing. It's crazy. We've got the version from Fusion here. This actually looks really good.

That is working perfectly. It's a little bit buggy when you can fly through the floor and stuff, which is weird, but overall, pretty good. Then, if we have a look at this version, this is from Fable 5. I would say it's not quite as good as Fusion.

So, Fusion won that round for sure, and then we have like a kind of vibe-coded version of kind of like Skyrim style called Frostfell, and you can see this is the output from Fusion, looking pretty good. Slightly buggy when you move around, but it feels fairly smooth. Big, fast, open world. Looks really cool.

So, that's from Fusion. If we have a look, this is the one from mixture of agents. I would say this one actually feels a lot nicer, a lot smoother when you move around. It's not buggy.

Looks pretty cool, and then this is the version from Fable 5, which is just a slightly bit buggy when you move around, but overall, pretty good. I wouldn't say this is interesting on the landscape, but I would say they all pretty much tie in terms of ... Let's try and go in the water here. So, that is a tie, and then we created a vibe-coded version of Doom here.

So, this is from Fusion. Pretty cool game. Fun to play. Easy to use.

Feels a little bit like an old-school Mega Drive game, something like that. Overall, it is a little bit buggy when things move around, but overall, good. This is the version from mixture of agents from Hermes, which is quite fun to play, easy to use as well, and now we have the version from Fable 5, but the problem here is we can't actually move forward. It won't allow me to move forward.

So, overall, I would say mixture of agents or Fusion won that. I'm going to put mixture of agents, but from what I'm seeing so far, Fable 5 won three rounds, the Crypt Racer game and WebOS. Hermes mixture of agents won three as well. Fusion came in last.

So, I mean, Fable 5 and mixture of agents seem to be the best. I would be very interested to test out mixture of agents with Fable 5. I think that could be absolutely insane. And the other thing I would say here is that if you're on the subscription plan with Claude, you're probably going to run out of tokens with Fable 5 pretty quickly.

So, I think that as a good backup, if you need really good outputs, then using Hermes mixture of agents as a backup when you run out of tokens on Fable 5 could be a really good option right there. So, that's the way that I would do it. I would use Fable 5 as a default. It's easier to go back and forth with Fable 5.

You can use the Claude CLI. It's easy to use. You've got everything stored inside the workspace in your agent OS. But then if you're using Hermes mixture of agents for those, you know, 10 or 15 minute prompts where you can walk away and come back later and store everything inside the workspace, I think that's a good backup option.

And bear in mind as well, the Fable 5 is only available on the subscription for until July the 7th. So, you know, at that point, you might switch to something like this. And I know that Hermes are working on Hermes Bench as well, which will be pretty interesting to see, okay, what do they get back and what can they achieve with cheaper models and which models perform the best depending on the setup. So, thanks for watching.

That's basically our system. If you want to get all of this plugged into one system with Hermes agent, Hermes mixture of agents over here, we have Claude plugged in with the CLI. We have all these amazing systems like an SEO agent, a video agent, a memory system as well. You can get that all inside the AI Profit Boardroom.

Link in the comments description or go to theaiprofitboardroom.com. If you want to compare everything that we've created side-by-side and see for yourself, well, number one, I would say, like don't listen to the benchmarks, just test it out yourself. And number two, if you want to see what I've created, you can check out GoldieBench. So, I hope to see you inside this community.

Link in the comments description or go to theaiprofitboardroom.com. And if you go to the new daily update section here, you can find our full system for the agent OS. You can see the last update date. We've got a video tutorial on it.

You can get the zip file for it. And every time something useful comes out like mixture of agents, we actually test it and we give you a full guide on how to use it plus a step-by-step roadmap. And we update this with daily tutorials. Inside the community, you can ask questions and I personally answer them with video tutorials.

Plus there's always people online 24 sevens, so you can get help and support whenever you need to. Inside the calendar, you can drop a weekly coaching course. And then inside the map, you can meet people in your local area who are building with Hermes and Claude and whatever else right there. So, I hope to see you inside there.

Cheers for watching.