Hermes Agent V0.19 Just Changed AI Agents Forever!

Hermes has a brand new update, Hermes Agent V 0.19, the quick silver release. I'm going to walk you through exactly how it works, what it means, how to use it, and how to get the most out of it. Plus there's two new pre models that you can use with Hermes agent. And I'm going to walk you through the quick silver command center and Hermes V 0.19.

This is the brand new release that just dropped. So that Hermes now replies, um, actually 80% faster agents finish jobs. Even if the app crashes and one gateway now runs a whole fleet of different agents. I updated mine the day it shipped.

I'm going to walk you through the whole walkthrough today and how it works directly. So how this works is basically we have the new quick silver core and V 0.19. So the first word in has a response time in less than one second. You can have multiple agents work walking through.

Now, what is the problem here? Well, basically with Hermes agent previously, imagine you have or hired the fastest assistant in the world, then making them sit in a waiting room before every single task. Well, that's what running AI agents felt like before this week. So before you would ask a question and stare at a blank screen, sometimes for like four to six seconds, I remember testing out grok 4.5 recently.

And it was super slow to reply. So you would hand an agent, a big job, close the laptop, come back to the job, didn't even work. And you would run one agent for everything. So the agent is super slow and every command also needed to be clicked, approved, like you were just babysitting an AI agent and adding this all up.

It made agents super slow and also not very useful. So the quick silver command center now breaks, breaks that cycle for good. And you might say, well, this sounds like another update. I'll need a weekend to learn.

The whole upgrade is super simple and easy. So what you can do is you can go over to your agent OS link in the comments, description, or go to the bottom if you want to get mine. And then from here, you can go over to the manage section and then literally you do is just click update Hermes right there and you can get instant access to. Now, if you're wondering what Hermes is, it's an open source AI agent from news research.

It lives on your computer. It talks to any AI model. You point it out, does it work? And V 0.19 is called the quick silver release.

And the name is literal because the team rebuilt how fast it wakes up, how fast it answers and how safely it can run many agents at once. So there's three ideas that cover the whole release. Number one answers instantly. So the time from pressing answer to seeing the burst were dropped by about 80%.

So cold start went from 4.3 seconds to 0.9. You never lose your work. So background jobs and final replies and now written into a database. So if the app or gateway crashes mid job, it finishes and delivers when it comes back.

And you can command a fleet, not a single agent. So one gateway can now route different channels to completely different agent profiles, each with its own model, memory skills, and secrets. So if you want to indicate how this works, well, you can type a message anywhere it could be in the desktop app, could be in the agent aware. So if you want the gateway checks, who should answer the call wakes up in under And cold start initialization dropped from 4.3 seconds to 0.9 seconds.

So if the model thinks before answering, you now watch it, actually thinking in real time, instead of just staring at nothing as well, which is useful. Also. It can delegate tasks to sub-agents now in a much more smarter way. And everything lands in your records for sessions as well, which makes it better than ever.

Now, if you want to indicate what the release notes, what are the main points This is not like some fancy new update where, you know, there's a bunch of new features, it's really focused on one thing, which is speed. And also there's two new, three updates for the model. So you can plug into this. So if you look at all of these updates, there's nothing like massively changing here in terms of what it can do.

It's really just about speeding up the agent, which I think is what the people want, to be honest with you. So this has become much better at delegating to sub-agents. And then the main feature here is it replies land 80% faster, which is fantastic. Also has smart approval.

So there's a second brain that checks like the risky stuff. So if you've ever like left an agent running and then, you know, you're worried about what it's going to do when it's running in the background, well, basically you can switch it on by default so that every basically flagged command gets approved by you manually, or you can switch on full auto. So you get a choice between like full auto mode or getting every approval, every command to getting flagged by hand. So you can choose between manual and auto approvals.

And there's also a third way now, which is it's on by default. So when a command gets flagged, a separate AI independently assesses it before it runs, which is great as well. And you can also set up your own deny rules as well that block stuff in auto mode as well. So if the agent wants to run a command, there's a smart gate where a second AI assesses it, or you can, it can run, or it can ask you as well.

And you might also say, well, letting agents just run commands in the background scares me, you know, good. I think everyone has that instinct. So the default is now second AI checking flagged commands plus deny rules. You write once the hold even in full auto mode.

So you get the speed of autonomy with breaks you control. Now also a single Hermes gateway can route messages to different profiles, which is new. And each profile is a fully separate agent. So for example, we can have like one gateway for our AI agents, and then it will actually delegate tasks to different profiles.

So we could have one profile for SEO, one for writing, one for ops, for example. And then these different AI agents can be delegated to using that system. We actually have a system for this where you can see here, if we go over to the chat section, we can have all of our different agent profiles set up, and then we can auto delegate from the main Hermes agent to those different profiles using this new system that just dropped, you might also say, well, I don't need multiple different AI agents, but if you've got different specializations or different tasks running in the background, it is good to have separate profiles. The other benefit is that you can delegate tasks to different agents with different APIs.

So for example, you could have one with HY3, which is a new free model on Hermes agent, and then you could also have another agent profile, and that could have Laguna S 2.1, another free model that you can use with Hermes agent. And then you could have a local model as well running. And so if you have three different free models, sometimes you're going to get rate limited or sometimes it's going to be slower, and so you can delegate your tasks automatically to that one main Hermes agent, and it will auto route the sub-agents directly from that one main agent, and you don't need to manage it yourself, which saves a lot of time. There's also a bunch of like smaller upgrades as well.

So it's got like reasoning dials. You can use these terminal commands. You also got stacked skills. You got real secrets management.

You got sessions that you can actually hand over to anyone. So you can export any session to mark down a HTML and a redact flag scrubs secrets before it leaves your machine. You've got a billing without the dashboard now as well. So you can run the command forward slash subscription.

There's also a panic button. So the safe mode now where you can boot Hermes with every customization off. If you need to fix something or if like the config on your Hermes agent is broken. And there's nicer buttons as well.

If you're using like telegram or discord or matrix, you actually get nicer buttons when you're using Hermes agent as well. Now, if you look at it, you know, we've got the old way versus the new way. So v0.18, the old version was like, you're waiting, you were babysitting your agent. You know, you had 4.3 seconds of cold startup time before the agent even wakes up.

And then you had background jobs, just breaking silently if the app crashed. You have one agent that answered every channel with the same brain. You had to approve risky commands by hand, one by one. You also had a full auto mode with no breaks at all.

And API keys were sitting in plain text config files and sharing a session meant copying and pasting and hoping that you kept the API details secret in that whole session. Now with the new way with v0.19, number one, it should reply in about one second. It's self-running. So you get a 0.9 second cold start.

So it also gives you it's thinking notes as it's loading. Also jobs and replies are written to a database. So even if there's a crash on your Hermes agent, it's still going to work in the background and one gateway routes each channel to its own specialist profile. The second AI can assess flagged commands automatically and your deny rules hold even in full auto mode.

Plus you've got keys that can be kept secret and you've got redact as well. So you can scrub the secrets on your exports. If you hand in a session to someone as well. Now, some people want to say, I'll wait until AI agents stabilize before I learn them, but you know, they're not going to stabilize like v0.19 shipped 2,245 commits in one single release.

So the people learning whilst it moves are the ones who will be unreachable when it settles. Other people will say, well, open source AI agents are just toys and it's in the paid ones, but honestly, like Hermes agent is way more powerful than any other agent that I know out there. It can do way more. You can customize it way more.

And also with it being open source, you can run it as a free models inside the AI agent, and then also the harness itself, Hermes agent itself is free and open source. So, you know, this is a really, really powerful model that is free to use as well. And also some people say, well, I don't need sub agents, but honestly, the sooner you learn how to use sub agents with AI, the faster you're going to move. And I'll show you some examples in a minute, right?

You also might think, okay, this stuff is too technical for the average person, but if you look at the AI Profit Boardroom and our testimonials and reviews, we've got 200 pages of wins of people growing, learning, using that agent operating systems and absolutely loving it, right? So if they can do it and I can do it and we're all non-technical, then you can do it too, right? You just have to get started. So let me show you some other examples of how this works.

So for example, we have the Kanban board over here and number one, our Kanban board is going to be way faster when we're delegating tasks, but number two, when we're building custom workflows like this, right, we're going to have multiple different AI agents and our Hermes agent can now delegate its tasks to multiple sub-agent profiles that we have created. So if we create like a video director, our main Hermes agent can now delegate tasks using the delegate tool to sub-agents in the background, and we don't need to babysit and manage everything. And also when you're running a board like this, where you, for example, creating SEO content, like you see in this blog post or a full video, as you can see here, when you're doing that sort of stuff, the main thing here is that you don't need to approve every single flag command because you've got that second AI now that can just assess it and go, yeah, it's fine, no problem. The other thing that I would say here as well is like, it's a good job that everything's getting faster because if you're using a memory system, well, what that means is it can retrieve memories and move faster now when it's pulling up personalized context.

So for example, we have this memory galaxy inside the agent OS, every single memory, every single action, every single task that I do with Hermes is actually useful and memorable. It will add inside this memory galaxy and now it can retrieve and use that even faster in the future. And then additionally, when we're using custom workflows, for example, like the Hermes Apollo system right here, it can, for example, use our builds, as you can see right here and pull them up faster so we can create full web pages with our voice, we can get it to operate our computer, we can create amazing stuff, it's saved inside our workspace, everything that we've spoken back and forth with Hermes Apollo about is all ready to go and this is a live voice agent that just works in the background. Same with Hermes Oracle and Hermes Astros, right?

These are powerful systems that can monitor our competitors, come up with content ideas and automate that for us. So if you want to get my full setup with all of this built in, you can get that inside the Air Profitable Boardroom, link in the comments description or go to theairprofitableboardroom.com. Inside the Air Profitable Boardroom, this is my AI automation community that helps you save time, grow and learn skill with AI automation. Inside the community, you can post questions, get help and support in real time.

I personally answer these questions every single day. Inside the classroom, you can actually get access to all of my best trainings and you can get our new daily updates, you can get our AgentOS system. We add new trainings all the time, so you can see all these new trainings we've added in July. And then also we have a bunch of step-by-step courses and guides you can get over here.

You can also jump on four weekly coaching calls where you can meet other people using Hermes Agent, ask questions about Hermes Agent, share your screen, et cetera. And then inside the map, you can meet people in your local area who are building with AI agents like you. So feel free to get out, link in the comments description or go to theairprofitableboardroom.com. Thanks for watching.

Bye bye.