Hermes just got 10x better (and FREE!)

**Julian Goldie** (0:00)
There's been some huge announcements that make Hermes 100 times better, and I'm gonna walk you through some of the latest Hermes agent updates and why this is so important. So number one, there's a new era of efficiency for your chat sessions in Hermes agent, which means that this can reduce the size on your disk by upwards of 78%, 78% smaller, and on average, 60% smaller to use, which means it's way faster to set up and way more optimized. Additionally, there are two free models inside Hermes agent. Number one is Laguna S 2.1. This is now free for two weeks on this portal. I'll show you how to set that up in a second. I'll show you some examples of what we've built with it.
And we also have HY3, which is your chat database just shrank by up to 78% using the new update from Hermes Agent and then additionally, you can now use HY3 and Laguna S 2.1 for free using Hermes Agent. How do you set up?
So what you can do is if you go to your manage section on your Agent OS, you can grab mine, link in the comments description, go to the model section over here, change the model and then you can select News Portal.
So News Portal here and you can use the free model. So you have Tencent HY3 and also Poolside Laguna S 2.1 for free directly with Hermes Agent. Now, if you want to get the latest version that's way faster and smaller to load, then you can actually update Hermes inside your manage section to go to manage inside Agent OS, then scroll down, click on update Hermes and from there, it can just automatically update to the latest version as you can see right here. So let's talk about the first one, your chat database shrinking up to 78 percent. Not very sexy, but it is big. Basically, Hermes stores every chat session in a local database and heavy users watch it balloon. The new optimization cuts it by 60 percent on average up to 78 percent. And if your database is under one gigabyte, the migration runs itself. So bigger ones get a prompt. Now, why should you care? This means faster session loads. It means snappier history search and a lighter desk disk footprint on the machine that your agent lives on 24 seven. So before your chat would be 100 percent, after you're saving about 60 percent. And you also got a bunch of one click free models that you can use with Hermes agent now too. So we have Laguna S21, Hy3, and Local from Olamma. These all plug into your Hermes agent, and then you can plug that into your agent operating system to build and automate anything you want. You might also say, OK, which model is the best? Is it Laguna? Is it Hy3? Is it Local? Here's what I would recommend.
So for us, when we were testing it out, we've built out of 50 different things with each of these models. So when we were testing out Laguna, this is more of a free local model, but you can get it on cloud as well for free on News Portal. So if we have a look at Laguna over here, you can see that it scored pretty average when it came to our builds with Hermes agent, especially on local. So it could build stuff, but not that beautiful, not that interested. I would actually say that out of all of those free models you can get access to, HY3 is your best option. It can create some pretty cool stuff, it can create some nice designs, et cetera.
Let me show you a few examples. So it can build like nice 3D games.
It was pretty cool building out these games like so as well. And it actually did a lot better just in general than anything else that we tested. HY3 is your best option with Hermes Agent. So for agent loops and coding, HY3 is performing the best. Hardest reasoning apparently is S 2.1.
But for my own tests, I would go with HY3 out of all of these.
Also, what you can do now, if you didn't already know, is you can use Cloud with Newspot with 2 So this is Cloud, Hermes agent living in Cloud. And you can basically set it up in like 60 seconds. You can sign in, just click through, it's literally just like click a couple of buttons. And once you've done that, you can run Hermes Cloud as an agent profile. And from what I tested, it's even faster. So when you combine Hermes agent and the new updates in terms of smaller database, three models, and then use Hermes Cloud as well, you know, you're absolutely going to crash. Also with the new dashboard updates, which I showed you before, like the manage section, for example, you can update in one single click and you get the models in pretty quickly. So before, if you're looking at like a couple of months ago with Hermes, getting an agent meant running a terminal, a Python environment, and even then you didn't get back. And then you'd be paying API rates to give it a good brain, enough to matter. And then also most people quit at stage one. I've seen loads of people quit and then just completely cancel because they couldn't get it working.