Hermes Agent Just Got 5X Faster!

There's a brand new update from Hermes that makes it way more powerful for agent swarms, for using different tools across all the parallel agents that you have, and this makes it way more powerful. So you can see the announcement here, it literally just dropped a few hours ago and it's called segmented tool batch dispatch. I'll explain what that means in a second, but essentially this is a more powerful version of the Hermes swarm system, which means your agents can run in parallel, and Hermes have just changed how your agent runs tool calls. So before they only went in parallel every single one of those tool calls were safe, now every safe call fires at the same instant, so only the ones that must wait wait, which means you can get way more done.

So you can ask and you can get multiple different tools running at the same time with Hermes agents. So for example you could give it one task and it can read a file, search, fetch, check calls and list all at the same time, which means number one is faster, and number two that your agent swarms work better together as well. So prior to today tool calls would only work in parallel if all were safe to parallelize. Now you gain big speed ups for parallel tool calls so long as any subset are parallelizable, according to Teck.

I'll explain in simple English what that means in a second, but here's the whole change. Before today Hermes would only run if all of them were safe to run together, if there was like one sort of a slightly questionable tool call inside the whole batch, then it all dropped back to each tool call being done in different steps and that means that your tool calls are way slower. So now it runs everything together and only holds ones that can't run, which means that you get a lot of big speed ups when it comes to using this system. So I would call this the one at a time problem and here's what this solves.

So picture asking your agent to do five things like check your calendar, read three files and pull the latest information in. None of them depend on each other, they could all happen at once, but your agent does them one at a time and then the first finishes and then the second finishes then the third and you sit and wait whilst it just plugs down the same list and so what happens with Hermes before is that five quick jobs would take like five times as long as one and the Hermes swarm system actually fixes that for good. So you can run multiple things at once. Now you might say okay that sounds like it could break something but if there's anything that shouldn't work or shouldn't be run inside the tool call system, that one is held back, the rest of the tool calls run at the same time and that means that you get the speed without the mess and here's an example.

So if you look in terms of speed, this would be the previous one at a time system and it would be super slow, this is if you're running everything at once. So it could be the difference between like 10 seconds if you've got two seconds per task versus two seconds because every task is running. It's the same five jobs but it's just the example of like a queue versus a swarm and that's really the difference between, you know imagine if you can move five times faster when it comes to building, when it comes to shipping. Well that's what you can do now because you've got this system.

Obviously tool calls don't take two seconds each but that's just an example of how it can run. So you want to think of your agent as a worker with a to-do list. The old way was a queue, you do one task, you finish it, you start the next, you finish that one, down the line and the line again. The new way is a swarm so every task that doesn't step on another one's toes goes out in the same seconds.

Reading a file doesn't hurt reading another one so both can go at once. The only thing that has to wait is a task that changes something the others are counting on. That one holds back, everything else already went. So you get the same list, the same correct result but a fraction of the wait.

So you can ask homies to do something and do several things at once. You know that could be for example like create an image, do a video, research this, read that file, edit this etc. And it can do it all at once and it looks at each call and asks one question. Is it safe to run alongside the others?

A read is safe, a write that the next step depends on is not. Every call that's safe fires in the same instance and the one call that must wait for you well that's just held back. So it's way faster as you ask. Homies will actually sort and filter through, the safe stuff fires, the risky stuff holds and then you get everything merged into one.

Runs way way quicker. Now we've actually got a system inside the AI Profitable Audio where we can have multiple different homies agent profiles running and then we also have for example all these different workflows with homies as well. And they all depend on tool calls. So for example if you look at Homies Oracle which pulls in the latest trending Twitter news, sorts it by popularity and what's trending the most.

You can see for example that something like this would use a lot of tool calls. So for example like we can publish content straight to our website, we can draft social media content. But if there's anything Homies doesn't like, before it would slow down the whole process and all our tasks would have to run one by one. With this system everything can run way faster, way easier, way more efficiently and that's exactly what we want.

Now if you think about okay what does this give you back? So the old rule would take five times longer versus a new rule if you had five tools that were being called at the same time. So it's just way faster, way smoother, way easier to use and also it makes using Homies agent way more enjoyable as well. You might also say well this is a tiny little dev tweak.

It won't change my day but multi-step jobs are most of what an agent actually does. So research, file editing, lookups, reports. You know if you cut the weight on every single one of them then the whole feel of working with an agent changes. You get faster answers and you get more stuff shipped.

It's the same for example if you're using a Kanban board. Like we've got this Kanban system inside the agent OS as well. So if we go down to the Kanban section here we can get more stuff done at the same time using the system because you can all loop around faster because nothing's being held back. So that's basically how it works and this is the whole method of the swarm system.

It's going to sort and look and filter through the tool cores. It's going to have the swarm running with every safe core. It's going to hold back the ones it doesn't like and it's going to merge all the results later. And that's basically how it runs step by step.

Now you also might say okay doesn't running something like this use a lot of tokens? But we actually have a token minimization playbook inside the AI Profitable Boardroom to reduce the amount of tokens you use. And also with Homies like they have a lot of free models and you know with those two systems you can reduce the amount of tokens that you use. You can even run this on free local models as well.

A lot of free local models run agentically now as well. A lot of people say well I'll wait until Homies is finally ready and then I'll start learning it but the people running agents in production now are improving every single day. If you wait you know you you don't start using agents you don't start learning them now you're just going to fall behind because this stuff is moving so quickly it's so powerful it's getting better and better every single day that if you don't start using agents now you're just going to fall behind. There's no way you can compete with someone who's who's moving this fast every single day.

Now if you want to get the full Homies swarm system built for you you know you can install Homies yourself and get the speed up free but you know if you want to get the whole thing done with all the workflows I've shown you today so for example Homies Apollo the voice agent, you've got Homies Oracle which pulls in the latest trending news from Homies. You've got for example Homies Astra so it can monitor keywords in your industry and then give you content ideas and you can plug that into our video agent inside the agent os as well. You can also get the memory system that plugs into all your agents and the Kanban setup and all these cool tools as you can see right here. If you want to get all of this system you can get it inside the airpuff warden along with a room of 3,900 operators running this exact setup and loads of trainings plus a coach coaching course four times a week where we can wire Homies together and just improve it together.

So what did you learn today? Well you learned what the one at a time problem is, how this new setup keeps your agents safe but at the same time makes them faster, how to build Homies agent swarms using the Kanban, how to move way faster with this sort of stuff and how to get it. By the way just run Homies update to get the latest version of this so that you can start using it and again you know I think with this you're going to stop waiting for one call at a time and you can just let your agents run all at the same time and that sort of thing. So if you want to get our full agent operating system set up and all our best trainings and Homies you can get that inside the airpuff warden.

Link in the comments description or go to the airpuffwarden.com. Inside the community you can ask questions, get help and support. I personally answer these with a video tutorial every single day so I answer each of these questions with a video tutorial and try and help you as much as I can. Plus there's always people online 24 7 to help you whenever you need it.

Inside the classroom you can get access to all of our best trainings, tutorials, lessons etc. Inside the new daily update section here you can find our agent OS system with the last update date, a video tutorial for using it plus all our best resources. If you're interested in Homies agent and agent operating systems we have a full one hour course on that right here as well. And if you want to learn how to reduce the amount of tokens you actually use you can use our token minimization playbook here.

Inside the calendar we have four weekly coach calls you can get help and support in real time. Inside the map you can meet people in your local area who are building with AI agents like you and this is all available inside the AI platform so feel free to get it and I'll see you inside there. Cheers for watching.