Solar Mini 4: FREE Hermes Agent Model!
AI News Today | Julian Goldie Podcast
October 8, 2026
Solar Mini 4: FREE Hermes Agent Model!
Speakers Julian Goldie
Julian Goldie (0:00)
Solar Mini 4 just dropped, and you can use it with Hermes Agent completely free right now. This brand new AI model is fast, powerful, and it's free for a limited time. So you want to grab it before the window closes. In this video, I'm going to show you how to plug it into Hermes Agent in just a few clicks. So you get your own lightning fast AI agent doing research and tasks for you all day, every day for free. Plus there's one thing you absolutely must do when you set up, get this wrong, and you'll end up on the paid version by accident. So stick with me to the end, and I'll show you exactly how to do it. Plus, I'll show you a whole list of other free models you can switch to anytime, which I'll reveal later. Let's get into it. So there's a brand new free model for two weeks on News Portal. It's called Solar Mini 4 from Upstage AI. So this is a new free model you can run with Hermes agent as 3 billion active parameters. So it's a mixture of experts model with 35 billion total, 500,000 token context window, artificial analysis score of 24 above models with 10 times the active parameters. So we'll get into how to use this and how to get access to it. Basically, what you can do here is you could connect this to a new portal, or I think you could use this with Hermes Cloud. Let's take a look over here.
So Hermes Cloud, basically, you can deploy from the cloud and you can have an instance in the cloud with Hermes agent and you can use the free models directly on them. Now, if you want to start testing that, basically, you can go to port.newsresearch.com/cloud.
And then from here, what we can do is we can create an agent. And if we have a look at the model section over here, we're just going to scroll down and then we are going to click on free. And you can see the Upstage Solo Mini 4 is available for free right now. You'll see all of these other models for free. So, Matron, Poolside, Inclusional, Poolside, Laguna, XS 2.1, Inclusional AI, Ling 3.1 Flash, Stepfun, Step 3.7 Flash, Ling 3 Flash, Matron, LongCat 2.0, and Upstage Solo Mini 4, which is the one that we shall test today. So, if you want to use this, let's go into Hermes Desktop. And then from here, let's create a new profile inside Hermes, which is going to call this the name of the model, which is Solo Mini 4 We'll create the profile. When you add a new profile, by the way, you can use an existing profile that you have. Now, what I like to do is have a different profile for each model, just so I can switch between them super easily and test them out. Then we can click on profile. Once we've done that, it's going to look like this. And so, what we need to do from here is set up the model inside the list. And we may need to just update this. We refresh the models here. You can see all the free models that we've got access to. Let's edit them. And then we've got Solar Mini 4 there.
And you can see we now have Solar Mini 4 working with Hermes Agent, actually pretty fast. So, I just posted the question, test. And it says, test and receive, session is live, clear heads up, blah, blah, blah, 100 memory is switched off. What would you like to do? So, for example, here, if we're like, okay, I want you to do the latest AI Automation news research on what happened in the last seven days in AI Automation.
Let's just see how that works. Click enter. So, and you can see this reply.
Now, this is not like frontier level model. So, for example, it's not something that you're going to use in the same way that we do something like Claude O plus 5.5. But for a free model, it's pretty fast to use. You can easily set it up inside Home is Desktop. You're good to go on that already. And you can see this come back to us with the research right here. So, for example, Oracle launched Fusion Claw. Didn't even know that, actually. That's pretty interesting. And then it breaks down the news. That is pretty good. Pretty good for a response. Pretty fast as well.
Super easy to set up. And it's got a breakdown of the sources, what stands out, et cetera. No problem at all. So it can handle your basic, agentic tasks. We can also use it inside Agent OS over here. So if we go to Home is, we can add that profile that we just had. So we can see Solo Mini 4 here. And then we can chat with Hermes directly. So, but this is pretty good. A lot of people say, oh, agents take a while to respond, blah, blah, blah. But from what I'm testing so far inside Hermes desktop, like it's really fast to apply. Pretty good. You can also add it if you go to your models inside your dashboards. If you run Hermes dashboard inside your terminal, you can open up the dashboard. Or if you've got the Agent OS, you can do this. And then if we go to the set main model section here, we type in Sola. You can see Sola Mini 4 is ready to go as a model to use right there.
4 more minutes of transcript below
Thousands of transcripts fetched by people building searchable podcast archives
Fetch the whole transcript
The demo key returns a sample episode in full, no card needed:
curl -H "x-api-key: pt_demo" \
https://spoken.md/transcripts/1000651996090Markdown with the speakers named, for your notes, your knowledge base, or anything that makes HTTP calls.
From $0.10 per transcript. No subscription. Credits never expire. Prices exclude VAT, added at checkout for EU customers. Not what you expected? Email us within 14 days with 20 or fewer credits used and we refund the pack in full.
Using your own key:
curl -H "x-api-key: YOUR_KEY" \
https://spoken.md/transcripts/1000793841948