Run Hermes Agent FREE With This NEW Model 🤯

**Julian Goldie** (0:00)
Free AI agents just got a whole lot more powerful with N2 and Hermes. So today, I want to show you how to plug a brand new free brain into Hermes Agent that turns it into a 24-7 worker building real stuff for you. I'll show you how to summon it up with one tap, switch between different agents whenever you want, and have them team up and pass jobs to each other automatically. And here's the best part, every single thing I show you today was built for free with this new brain. Stick with me until the end, because there's one simple thing you have to do to unlock it, and I promise it will change how you use AI forever. Let's lock in and get into it. There's a brand new free API you can plug into Hermes Agent. You can see some stuff that we've built with it today. And this was all powered by something called N2 Pro, which is a new free model. It's actually built on the Qwen 3.5 architecture. And you can grab this free API and then plug it into Hermes Agent. So for example, if we go to N2, which is a separate agent profile just built for the N2 API with Hermes Agent, we can use it and we can use this API for free. So how do you set it up and what are the benefits? So basically, if you want to set this up quickly, you can just go to manage, go to models, then you're going to change the model on here. So you would just go to change. And the benefit of this, of course, is that you get a free API. So you can use the most powerful agent in the world, but you can use it for free on this API. So you just go to OpenRouter from here, and then just make sure that you are using the free API, which is N. And you can see the details of the model right here. Now, again, this is called N2 Free. So that's the model you want to use.
If you want to see the model weights, you can see they're available on Hug and Face right here. Now, this is designed as an agentic model with agentic thinking, which is why it's great for Hermes, which is an AI agent. And so this is basically a really powerful way to use a free model with a Hermes agent. And it can build some pretty cool stuff, which I'll show you in a second. What we've actually done is if we go to our Hermes profiles, for example, we've set up a separate profile for using N2. So if we scroll down here, and this is a cool way to use it because you can easily switch between APIs based on the agent profile for Hermes. So for example, we have one which is using step 3.7 Flash as a free API. And that's another free model on News Portal. But then we have N2, which you can see right here. And this is using the N2 model. And so you can have different profiles switching between different APIs. And then we can chat between each of these agents as you can see right here, which is pretty amazing in itself.
Now, the other cool thing about this as well is like if you're trying to create sub-agents, you could use the free Kanban board from Hermes Agent. And then you could have one profile, which we've got here for N2, and then one profile with a normal API, and you could switch between them and they can delegate tasks. You can plug in a task over here, it will triage it and then move across the to-do list using the different profiles and it will sign tasks and delegate tasks to different profiles. Now, is this all perfect? I wouldn't say so because of course, you know, if you look at the context, it's 262K, which is good, but it's not as high as, for example, like Claude. And I wouldn't expect Claude level outputs or like Fable 5 level outputs on this model. But is it good for agentic tasks? Is it good for subagents? Is it good for getting basic things done, which, you know, most people are wanting? Then, yeah, absolutely, absolutely. And so this is an agentic mixture of experts model from NexAgi, 17 billion parameters out of 397 billion parameters total. Built on Qwen 3.5 architecture, it accepts text and image inputs as well, which is pretty cool. And it's also used and designed for coding, tool use, deep research and long horizon agentic workflows. Now, what that actually means is that, for example, if you wanted to do long horizon on this, is you could actually go to the goal mode inside Hermes Agent, and then you could give it a task and you can just go off and work on that task for hours, like you see with these previous projects. So, for example, if you had something big like building a website, you could do that. And so we added this brand new agent to Hermes N2, and it reasons like a top model. It codes, it builds, it's free and it's free through OpenRouter. And you can summon it as a profile, which is what we like as well. So that means you can just quickly switch to N2 if you're like, oh, do you know what? I want to use this API as my main Hermes one, but then I want to switch to this agent profile over here for N2. Perfect. That's how it can work.