Hermes Agent V0.20 Just Changed AI Agents Forever!

Hermes Agent V0.20 The Herald release just dropped today and we're going to walk you through exactly what it means, how it works, how to get it and how to use it. So the easiest way to update is you can go to the manage section inside the Agent OS and then click on update and then you can get the latest version so we can update to the latest version right there. If you're wondering what the highlights are or how it works etc. So the first thing that you can do is you can now talk to Hermes which means that the voice speaks whilst the response is in the chat. It basically responds instantly whilst you're talking. It's got cited sources, agent to agent. It's also a lot faster. So first token in 0.9 seconds instead of 4.3 seconds and also a lot faster on Hermes Desktop. Also a Hermes desktop platform so you can now get artifacts with Sandbox Live Preview, Smart Approvals, CLI Power etc. Now we're going to run through the changelog and I'll show you this in more depth plus some examples of how it works etc. But if you're wondering okay what is the Herald release? Well here's a few headlines. Basically it's just shipped a big update and your AI can now finally hold a conversation. You can talk to it, answers as it thinks and you can cut it off mid-sentence just by speaking. Kind of like a phone call and also you can say your own wake work from across the room and your computer starts listening. So for example if we go over to the agent OS over here and then we go to the chat section. You can trigger this on and say hey Hermes how are you today? What time is it? And you can see that actually goes inside the chat so it can start responding to us directly from this. So it actually listens to us with the wake word and it's ready to go and then it can respond in real time. So for example we said hey Hermes what time is it? And then it responded the other day. This update actually dropped the other day but basically this new update is a bundle of all of these updates together. Also backs research with sources so you can actually check, ping your other systems the moment the work finishes and it even talks to other agents as well. So what is the talk to Hermes update? Basically Hermes answers whilst it thinks and you can interrupt it just by talking. So it can stop, listen and adjust as you go along. So it's kind of like having a phone call with your AI agent. So the easiest way that I find to use this is I've got Hermes Apollo set up here and we can actually speak to Hermes agent in real time and then it can respond to us. So you see how we can speak to it directly here and then it will listen and then if we ask it for a response it will talk back. This actually has a voice as well but I won't play the voice on the playbacks. It just won't sound that good. But you can see how it's going back and forth with us in real time right here and we can interrupt it whenever we need to as well. With the wake word basically this is just hands-free control so you can have Hermes sitting in the background. You could have for example Hermes desktop on a different monitor or the Hermes agent OS and then what this means essentially is it would listen and when you say for example hey Hermes it will wake up and start responding. There's also voice on every platform now so you can send a voice note to Hermes on WhatsApp, on Feishu, Ding Talk, Line, QQ, etc. And it will automatically reply with TTS so it's got auto TTS replies and it's platform aware. Additionally there's a new grounded citation skill. So the grounded citation skill allows Hermes to produce research where it has sources for everything that's used in. So this is kind of like deep research where you just avoid hallucinations so you get better research back because it's citing the sources on every response. Kind of like perplexity does as well. Now inside Hermes desktop as well you've got the artifacts feature. So here you can view everything that you've created and used with Hermes agent across all your platforms. So for example we were using Hicksfield the other day and with the MCP with Hermes and Hicksfield linked it creates some nice images. Now all of those artifacts are actually displayed so we can view them over here as you can see which is pretty cool. And it's sandboxed live. So if you're running like HTML apps and you want to test them out etc. Then you can preview them inside the artifacts feature here and it's just a lot safer because they're sandboxed. Now something else that's pretty interesting is Hermes speaks agent to agent now A2A. And this is a new plugin that implements the agent to agent protocol. Which means that Hermes can discover, it can talk to, and it can be driven by other agent to agent compatible agents. Which is pretty useful. So one way that I usually use this already is we have the agent OS and then we have the agent mastermind here. So if I just post a message like hey hey you can see that all of my agents including Hermes can reply to us. And then we can drive our agents and they can talk to each other, bounce ideas off each other, and even add those ideas to our idea pipeline over here using this system. So it's super useful because if your agents can talk to each other, well then all of your workflows across all of your agents and all of the skills and APIs you plugged into each one can be driven so that Hermes can actually orchestrate them. Or you can get another agent to orchestrate Hermes. There's all sorts of like technical fixes like CLI, power, use a wave, correct the agent mid-turn and redirect it. And if Hermes is actually going the wrong way, you don't have to use forward slash stop and re-explain. You can just type a correction whilst it's working and the active turn is redirected. So what that means essentially is like you can steer it in real time. So whilst you're waiting for it to respond to you, if you see it's moving in the right direction, you can just say hey do this, do that. And you don't need to use any technical commands to steer it. It will just automatically pick that up. There's also some new improvements on compression, which means the recent conversation always survives. It's progress aware as well. And when it compresses a conversation, most of the context should show up in a better way. And the smarter approvals now. So if it's about to do something that it's not sure about, it will use the smart approvals just to make sure it actually gets it right and approve it with you before it takes any action. Like for example, deleting a file or that sort of thing. And additionally, Hermes is now faster. So on the telemetry gate, it's now 54 times faster and just way quicker to respond. This is also super useful with Hermes agent because if you're using something like a tool use, like for example earlier today, we said forward slash learn and gave it access to a guide. And then it learned the guide as skill MD file and replied. With DeepSeq particularly, it is really, really fast to reply. Like something that would have taken a few minutes with other models can now be done in seconds with this system here. So number one, it's faster to reply when you use it. But number two, if you're using newer models like DeepSea Fee for Flash, it just came out a few days ago, it's even faster. So both of those combined make it even better. Then there's a bunch of voice updates, text-to-speech improvements, core agent and architecture improvements. But that's basically it for the whole update. The one thing that I'll say here is like the desktop is improving a lot. So it's become a lot faster as well and it just performs better. It's also a lot nicer when you open up now. It didn't look that good the first time I used it, but it's getting better and better, which is great. And I think one thing to pay attention to from this whole update is the agent-to-agent protocol. Because if that keeps getting better, it's going to be way easier to control all your AI agents. There's also a bunch of security updates, which I think is needed because it's getting more and more powerful. And a bunch of bug fixes as well. With 647 contributors on this release. So that's basically it for the Herald release. You now know the key updates. Talk to Hermes. Wake words. Voice on every platform. Research is cited now. Outbound webhooks. Desktop is better for artifacts. It's faster to use. It speaks agent-to-agent. And there's a bunch of technical movements. And also, you can now correct it mid-turn and redirect it without having to use forward slash stop and re-explaining it. So if you want to get a full agent operating system for Hermes, where you can plug in your Hermes, you get custom workflows. Like, for example, we've got the chat, but we've got different profiles here. So we can test out new models. With Hermes, we have the voice agent that can build in real time. It can control our browser and our computer. And it responds live. We also have Hermes Oracle, which pulls in the latest news from our industry. And then we can automate content for social media and SEO directly. We have an outreach tool with Hermes agent that can generate leads on the spot. Then send email campaigns, manage your inbox, and give us the key stats of that whole campaign. And also, we have the studio where we can generate videos, images, and voice and preview them like so. Plus, we have our whole memory system plugged in, which means that we can give context to Hermes and all our other agents. And this is automatically updated with Hermes directly. So feel free to get that link in the comments description or just go to the AI Profit Boardroom to get access to our Hermes agent operating system. If you go to the classroom here, then go to the new daily update section. You can find the agent OS. And this is my community to help you learn, grow, and scale with AI automation. We add new tutorials. We've got loads of tutorials, for example, on Hermes and Buzz, how to use the Hermes learn feature, how to use mixture of agents, et cetera. Inside the community, you can ask questions and get help and support in real time. I personally answer these questions every single day of a video tutorial. And then inside the calendar, you can jump a weekly coaching calls, ask questions, share a screen, meet other members. Inside the map, you can meet people locally using agents like Hermes. So feel free to get that link in the comments description or just go to the AIprofitborn.com. Thanks for watching.