Olares recently sent their Olares One to try out. The Olares One is a compact, high-performance personal AI cloud device. It is designed to run heavy AI workloads, local LLMs, and image/video generation pipelines locally without sending data to external cloud servers. Oh, and it runs an open source Linux-based (Ubuntu-based) operating system!
0:00 Intro 1:25 Hardware specs 4:28 Connectivity and ports 5:46 Setup 7:56 Navigating to the Olares desktop 10:32 The Market app 11:00 Installing Ollama and Qwen 16:21 Installing Open WebUI 19:56 Installing ComfyUI 26:01 Installing Hermes Agent 30:34 Updating the Olares OS to get new apps available 31:48 Installing Router 34:20 Installing Lares 36:27 Other Apps in the Market 37:29 Final thoughts
Your support is very much appreciated. Thanks, guys!
AI Generated·May contain errors
Today, we're taking a look at the Olaris One,
which is a powerful personal AI server designed
to bring your digital life, applications,
and AI tools together all in one place. The good folks
at Olaris were kind enough to send me this unit to review it,
so I want to thank them for that.
The idea behind the Olaris is pretty interesting.
Instead of relying on a collection of cloud
services, subscriptions, third-party platforms,
you can actually run your own private cloud
Right from home.
In particular, this device is especially useful for
Running your own local AI models.
In fact, I would label this device really as an AI server.
But despite the focus on AI,
the Olaris One can be used for a variety of other tasks.
Like any computer, it can be used to host
your own web applications, manage files, set
up a media server, build your personal home lab.
Whatever the task, the Olaris One
is probably going to be well suited for that.
It's a very beefy machine in terms of specs,
but is it really worth the investment?
Let's go ahead and talk about that up front.
This machine retails for around $5,000 US.
Now for that price, you get some very nice hardware
with this machine, but how capable is that hardware?
How easy is it to set up?
How does the software deliver on its promises?
Well, let's take a closer look at the Olaris One
and see what this little machine can actually do.
So let's start with the hardware.
The Olaris One packs some seriously impressive
specifications into a surprisingly compact
So if I scroll down here on the Olaris One product page,
you can actually get a nice
overview of some of the hardware specs here.
At its heart is an Intel Core Ultra 9275HX processor,
which is a 24-core processor capable
of boosting up to 5.4 gigahertz for your graphics and AI acceleration.
Of course, you've got a very beefy GPU.
You've got the NVIDIA GeForce RTX 5090, and that comes with 24 gig.
of GDDR7 VRAM. It comes equipped with 96 gigabytes
of DDR5 memory running at 5600 megahertz
and a two terabyte NVMe drive for storage.
All of this is housed in an aluminum chassis
measuring approximately 12.6 by 7.8 inches
and just over two inches thick,
weighing about 4.7 pounds or 2.2 kilograms. In other words,
This isn't your typical mini PC.
It's essentially a compact workstation built around
a very powerful mobile GPU with plenty
of RAM, plenty of processing power
to handle all those local AI workloads.
Now, normally I would show you the actual computer itself.
I would have it in my hands and I would show it to you.
Now, why don't I have the computer with me
is because I don't actually have the Olaris
one with me here at my recording studio.
It's actually plugged up and running.
At my house, which is about a 15 minute drive from here,
because the Olaris One, it's not
really a typical PC.
It's really a server.
It's designed to be run headless,
a headless computer, meaning you don't typically plug
in a monitor and a keyboard
and run it like your typical workstation, right?
A server is designed to be run headless.
It can be plugged up anywhere in the world,
and then you connect to it from any other
device.
In this case, I would connect to it here on my web browser.
Here using the Brave browser on my machine here at the office,
I can actually log in.
to my Oleris One.
And this is the way you connect to the Oleris One
and install applications and run applications.
You do it from a browser.
You can log in from any machine using a web browser.
And if I want to navigate the desktop here,
you can see it looks like a standard kind of Linux desktop.
We should talk about the operating system.
The operating system is a Linux-based operating system.
And the actual desktop environment here
kind of resembles your
Standard like GNOME desktop environments and stuff like that, right?
You got a dock on the left-hand side
where you've got pinned applications.
You've got an overview of all
the applications that are installed as well.
Then you've got date and time down here in the bottom right,
as well as some system monitoring
information.
So really nice looking, very kind of minimal desktop, but I like it.
I think it's very clean and very beautiful.
Now, before demonstrating some of the software,
we should talk a little bit
about connectivity.
The Alaris one keeps things relatively simple in terms of connectivity
and ports. It only has four primary IO ports on the device.
You've got a Thunderbolt port
and HDMI port, a USB-A port, and a 2.5 gigabit Ethernet port.
And that's fairly minimal. That's
a fairly minimal selection of ports.
Now, because it's primarily a server, you're not connecting a
Not going to be a big deal for most people,
but if you were planning on using this device
as a workstation, meaning you were going to plug in a monitor
and a keyboard and possibly
other peripherals.
Then you're probably going to need some kind of docking station.
Now, out of the box, when I got my device,
I don't know if this normally ships with it
or if this was something that the good folks
at Alaris just gave me for purposes of testing,
but they gave me this very nice little USB hub here.
You can see I got four USB ports here.
On the back here, I've got a couple of extra HDMI connections,
SD card readers as well,
and then I've also got another Ethernet connection
as well all on that station.
So that's something that you probably are going
to need if you plan on running the Olaris one
as a workstation. So let's talk about setup and installation.
So when you first unbox the Olaris
to get up and running, all you need
to do is you need to plug in the power cord
and turn the device on and then plug in the ethernet
because you're going to need
internet. After that, you need a mobile device.
You need, you know, either a
Apple or Android phone,
and then you need to install this particular application, the
Larapass application.
And what Larapass does, it connects to your Olaris One device.
And for example, when you are very first setting up the device,
it actually installs the operating
system and everything, and it sets that up for you.
And it's very easy.
All it does when you first get into the app,
it scans for your device, connects to that
that device and then ask you, do you want to?
Install Alaris on your Alaris One.
You hit the install button, right?
It's very easy.
Other than that, you set up an account.
You're going to need a username and a password.
It's also going to have a vault in here,
a vault where it occasionally will generate
like a two-factor authentication kind of password.
So you can actually log in to your Alaris One
from any computer in the world.
You just go get your...
one-time password from the app, from the Larapass app,
and then you can actually get into the
operating system. One thing I noticed
when I first plugged in the device and powered it up,
I couldn't tell if the computer powered up or not,
because this computer is completely silent,
right? It's whisper quiet operation,
because looking at some of their specs, they claim that
that it runs under 23 decibels at idle and 38 decibels under full load.
I couldn't even tell it's running.
I couldn't hear anything. I put my hand on the device. I
couldn't feel any kind of vibrations.
So this is one of those, if you're worried about noise,
because obviously I am a video content creator.
I can't have really loud devices and loud fans
and loud clinking and clinking going on or whatever.
This device is super quiet.
So jumping back into the software,
let's talk a little bit about the operating system.
When you first try to log into your Olaris desktop,
of course, you'll go to a web browser
and you'll type a certain URL for you.
In my case, I've got desktop dot
and then my Olaris username dot Olaris dot com here.
And then I can get to my desktop.
The very first time you log in, though,
you may be greeted with a screen that looks something
like this where it's looking for a one time password.
Basically, again,
You can think of it as a two-factor authentication.
You get that from the LeraPass app.
You go get your password, and then you type that in,
and then you should be granted access
to your Oleris One desktop.
Now, Olaris OS is an open-source Linux-based
personal cloud operating system built on Kubernetes.
Now, do you need to know anything about Linux
or Kubernetes to really run this thing?
No, it's designed to be dead simple,
very easy to navigate UI, very easy app installations
and everything.
You do not need to be some kind of system admin
or network admin in the Linux world
to actually run this thing.
It's very user-friendly.
Finally, Polaris does include
a standard app store. If I click on this particular icon in the dock,
this is the market app. This is
your Olaris app marketplace where you can quickly
install various applications on Olaris. I'm not
exactly sure the total number of apps in the marketplace,
probably a couple hundred or so,
and you've got everything from local AI tools
to image generation software, media server software,
Software password managers, file storage apps,
automation platforms, now all of these applications
Applications run in isolated environments,
which helps keep these surfaces separate and secure.
There's also remote access through the Olaris ecosystem,
allowing you to reach your files and applications from other devices.
And what's interesting about this, reaching files from other devices.
If I open the file manager.
So this is a standard kind of file manager
but if I opened my file manager here on my workstation
here at the office this is my file manager here let me just make
a file I'm gonna go to file
and let's just create new empty file I'll call
it test.txt I just created this empty file
if I drag it over here to Oleris
It is actually copying that empty file over.
How cool is that?
So it's very easy to move files from the Alaris desktop
over to your local machines and vice versa.
Now let me close out the file manager
and let's go back to the app marketplace here.
Now,
I've been playing around with this for a few days.
I had a lot of applications installed that I've
been testing out, playing around with.
Right now, though, I am back to pretty much a base install
because I uninstalled everything
I had been using and I'm going to install.
All of that stuff back onto the system today with you on camera
and maybe give you a brief
overview of how some of these applications work as well.
So let me get out of this little application overview here.
Now, the very first thing I wanted to install
is I wanted to install Ollama because I.
When it comes to AI and my workflow,
I use a lot of large language models, and if I want to do
local large language models,
Ollama is kind of the application to install. So let's go see if
we can find Ollama here.
Now we got some categories that we could play around with, maybe AI.
It's probably in this particular category.
If I click on that, let me scroll down here.
Do I see Ollama?
I see Ollama Engine Base.
So that's probably something that's already here
out of the box.
So I may not even need to install Ollama
if it's already here.
If I search for Ollama, if I can spell it correctly.
Yeah, I mean, we've got everything here.
I think I just need to go ahead and install a model.
So let's go ahead and search for a model here.
So I know I want Quinn,
and I want 3.5-9b, and here we go.
Let's go ahead and get that,
and then click install, and it's going to download.
Now, these large language models,
some of these are very large downloads, so this could
take a few minutes.
Pause the video and I will be back once the installation
of QUIN 3.59b has finished.
And the installation of that particular
large language model has completed.
That took about 20 minutes or so to download.
Again, these AI models, some of them are very large
downloads and that really is the biggest part of setup
for a lot of these initial applications
is you're gonna have some big downloads,
but of course once they're there,
They're there, right?
So let's go ahead
and see if we can actually get Ollama now running with Quinn 3.59b.
So let's go ahead and if I actually go into the launcher here,
I've got Quinn 3.59b.
If I click on it, does that actually do anything?
Just.
Gives me some kind of information,
status information, which is kind of cool.
Opens in a different browser tab.
That was actually kind of neat,
but I was hoping just to get like a terminal with Ollama.
I mean, I don't have any way to launch Ollama here.
So I do know the Ollama engine base was installed
because we saw that, right?
So let me get back out of the dashboard here.
Let me go back into the marketplace
and search for Ollama engine base view.
So we don't have a button that says get or install,
but we do have create, and you see zero instances created.
So let's create a new app name.
The title cannot be empty,
so I'm just gonna call it olama, create,
and configure environment variables for olama.
So model source is going to be quinn,
if I can type it, 3.5 colon nine b.
Now it says it must follow this pattern,
olama colon slash slash quen 3.5 colon 9b,
and then model name, I'm just gonna call it
quen 3.5 colon 9b.
Model mode, it says chat or embedding.
I'm just gonna choose chat here.
Model supports, we've got vision tools, thinking, none.
Choose vision here, and then.
Next up, engine args, do we need engine args?
I'll just ask ChatGPT what to enter here,
but I added some engine arguments here.
And then, OLAMA require GPU memory,
so I'm sure there's probably a default size somewhere,
but I'm just gonna say eight gigabytes.
It says value must match this pattern here.
So it looks like it didn't like 8G,
it wants 8GI for eight gigabytes.
Let's confirm all of that.
And it looks like it is creating the Ollama engine base here
or the instance that we need.
And if I go back to our little application
overview you see I've got an Ollama icon here
and you see it's got a little timer there on it.
Of course that means of course it's being installed
or set up here so we're going to have to wait
a couple of minutes here. By the way when I was
This Olamic instance, I just noticed
that if I'd actually read the description here,
everything that I just filled out,
it kind of gave you exactly what needed to be there
for engine args and everything.
So, yeah.
Kind of cool.
So I was actually looking over to my far left monitor.
I have the triple monitor system here,
and I asked ChatGPT, and it spit out all of that
information.
I'm assuming ChatGPT just got
that from the Olaris 1 documentation, and it looks like
it finished creating that Ollama instance.
So let me close the marketplace here
and go back to our application launcher here, and
now I have Ollama, and now let's launch Ollama.
It opens a tab here.
It looks like a QIN 3.59b is installing
because we've got some status information here.
So I guess there's still possibly some setup going on here.
And it looks like it finished setting everything up.
So yeah, when you click Ollama now in the dashboard.
Yeah, you just get like this configuration
and status window basically telling you Ollama's running,
but it doesn't open like a proper terminal interface.
Although I think most people probably wouldn't want
to run a terminal interface here.
They probably would want a graphical interface.
So let's actually go to the market and actually install OpenWebUI.
Here is OpenWebUI. This is a graphical front end to Ollama.
So let's go ahead and get
And install, and again, it'll take a couple of minutes
for that to install, although it looks like
it's downloading pretty fast.
This is not as big a download as one of the
large language models.
And open web UI, finished installing.
Let's go back to the launcher, and now let's open...
The open web UI interface here for our graphical way
to chat with some of these large language
models and we get started so we have to create an account here.
I will call my user DT and then
I need to enter an email address here
and then I need to create a strong and complicated password
here. Let's go ahead and create our admin account here.
And then we get some information about the latest release.
All right.
And then we get to chat here.
First, we need to select a model.
And if I select models, there are no models available.
So one of the things about all of these applications
is because they're all in their own separate container,
you do have to connect them to each other.
So we downloaded the QIN 3.5 large language model.
But Open Web UI does not see it.
How do we actually get that connection to work?
Well, if I go down here and go to Settings, and then let's go to...
Let's go to Connections, and then Manage Connections here.
We have Ollama API, and we could click the Add Connection button here,
and then I need to add a connection.
But I don't exactly know the URL of the connection I need.
Where would I go and get that?
Well, I've got to go into the Olaris settings, which is this icon here.
This is the settings for Olaris, and let's go to Applications.
And then let me go and find, in this case,
I need to find Ollama, which it's running.
And then I need to go down to Shared Entrance,
click on that endpoint.
There is a URL here.
Let me copy that.
And then let me get that window out of the way
and let's actually add this URL here.
And then let me verify the connection actually works.
Server connection verified, okay, well I'll save it then.
And then let's go ahead and save all of this
and then get back out of this.
And now when I go and select a model,
I now have QIN 3.5 colon nine B here.
So let's actually test it out.
Write me a haiku if I can type correctly.
And it's taking a few seconds to think.
Here's a haiku for you.
Silent code speaks deep.
Digital dreams bloom in night.
Nature meets machine.
Not my favorite haiku, but it's not bad.
At least it did what I asked.
So that's a quick demonstration of how to get up
and running with some of these large
language models using Ollama.
Then install whatever.
Large language models you need.
Again, they're big downloads.
You're probably going to need multiple models.
I'm just not going to use Quinn 3.5.
I probably need to install some other models as well.
It's going to take some time to download a lot of that stuff.
Then you've got Open UI as your graphical interface to use that.
But what else could we do?
What other kind of AI tools could we use?
Let's actually go back to the market.
One of the things I've been playing around
with a little bit lately is Comfy UI,
which allows you to do image generation.
Using AI.
So let's go ahead, click get, click install.
This could take a couple of minutes to install.
And it looks like it finished installing.
Let's close the market and go into the menu system
and we have Comfy UI.
It says Comfy UI currently unavailable.
Maybe I needed to click the launcher here.
Yeah, click the launcher.
Let's click start.
And you notice this opened in its own tab, right?
There is the Olaris desktop.
There is Comfy UI Launcher in its own separate tab,
and we're waiting for it to...
Go ahead and launch Comfy UI for us here.
All right, and it says it has started.
You see we've got open, restart, and stop.
Let's open, and now it opens a new tab
with the proper Comfy application here.
And Comfy has launched.
Let's go ahead and play around with something.
So let's do an image,
and we could actually search for some kind of image.
Let's do, because I love Linux, let's do penguin.
Is there any penguin templates?
No.
Well, that's unfortunate.
How about horse?
No.
How about statue?
There we go. We found a statue of a man riding a horse.
I guess the horse didn't count when we were searching.
Well, let's go ahead and jump into this. So as soon as I launch this,
if this is the very first time you've seen comfy UI, this is a
graphical front end to using some of these AI models to create images.
But you have to install various models for these templates to work.
You can see I've got four issues found.
If I view details, it's going to tell me,
hey, you've got to install some stuff
before you can actually do what you're trying to do
with this image.
And you see I've got download,
and then I've also got server download.
Now, if you click download,
it's gonna download to your local machine.
So in this case, my office workstation.
I don't want those models on this machine.
I want it actually on the Olaris One server.
So I'm going to
Click server, download.
If I click this one, I think it will download all of them.
Let's see.
Let's go ahead, start download.
It's actually told me exactly where on the Alaris server
it's gonna download all of this stuff to.
So maybe I've gotta do them one at a time.
Yeah, start download.
And then this one, start download.
And then this one, start download.
It is verifying exactly where it's going to put it
on the Olaris one as far as what directories
these models are getting downloaded to.
And if you are somebody that actually knows
where these things belong, it's good to verify it
just so it doesn't put something in the wrong place
because if it puts it in a wrong folder,
then this is obviously not going to work.
And those four missing models have now downloaded.
And if I click the refresh button here,
Let's see if all four errors now go away.
Oh, one missing model.
So even though we downloaded this one,
it's still an error.
So this model might have been downloaded in the wrong place.
So let me go back to the Alaris desktop.
Let's go into the file manager here
and let's actually go into data here
and comfy UI.
So this is where all the comfy UI stuff should be downloaded.
Comfy UI.
Comfy UI. We just keep going down.
Comfy UI models, I think, is what we need to be searching
for here. Let's search alphabetically. Models.
One thing we need is we need a folder here called
text encoders. We have that.
And that is where that missing module should have been downloaded
to. So let me go back up a folder to models and go to checkpoints.
And yeah, here is the file that is the problem.
Let's cut, and let's go up to models,
let's go back up to text encoders,
and then let's paste, and I think that was the error.
If I go back to Comfy UI, refresh,
and see if the missing model now goes away, and it did.
All right, and let's collapse that sidebar,
and let's just run this template out of the box here.
There's one active, so it's running a job here.
And if you're new to Comfy UI,
and I'm pretty new to this, it's a node-based AI kind of
workflow where you describe exactly what kind
of image you want to build here.
And, you know, you've got various parameters
that you could work with here.
So you can see total 12%, it is.
Slowly generating this image for us.
All right, and it generated this image here.
If I click on it, yeah, neat image.
Now, how could I change this image?
Well, I could go back in here.
Let me collapse that sidebar and let me zoom in here.
Let's go into the text here.
So what are we doing here?
We're doing a closeup of a medieval knight
from the chest up alongside his barded war horse.
Let's make this a barded war.
Camel, right?
Let's actually search for horse elsewhere in this document.
Yeah, we're going to change horse to camel.
He's going to be riding some kind of armored camel now, right?
So now, well, actually, let's just run that simple change,
and let's see if this makes a difference.
And you'll see, basically,
how you can change things on this image generation tool,
this UI tool here, by just changing the description.
And it finished generating that image.
Oh, I like that.
That is a nice looking camel here.
Let's see.
Click on this to get a bigger preview.
Yeah, that is a very interesting image right there.
So let me close out a comfy UI here.
Let's go ahead and close that tab.
I can go ahead and stop it from running in the background too,
because I don't think I'm going to come back to it anytime soon.
And let's go ahead and try one more installation.
I do want to go back to the marketplace.
And one other tool that I'm not that familiar with,
but it is kind of neat, is Hermes Agent.
What is Hermes Agent?
Well, it's kind of like a chat AI assistant.
If I read the description down here,
it creates skills from experience.
It builds them during use.
It nudges itself to persist knowledge.
So basically a lot of AI chat sessions,
they only know what you've.
Told them during that one session.
This one, it kind of persists, right?
The past conversations,
you build on those past conversations with a tool like Hermes Agent.
So let me go ahead and click on Get.
And install, now the good thing about this
is a lot of times after installation,
you also need some large language models
installed for this.
And the one that I was going to install for Hermes Agent
was QUIN 359B.
So we've already got that installed
because we also installed it for using
with Ollama and Open Web UI.
So we've got Hermes Agent now up and running.
So now let's go ahead.
And launch it, so let me search here for Hermes.
Hermes CLI, Hermes Dashboard.
Let's try Hermes CLI, see what happens here.
And we get a terminal, and from here,
you would first have to run through a setup,
so Hermes Setup.
And here in the terminal, it's gonna ask you
a whole bunch of questions.
Now for me, I'm going to go through a quick setup here.
So let's just go ahead and run the quick setup.
And let me go ahead and choose terminal back in.
That should probably be local.
And connect a messaging platform.
I'm gonna skip that.
That could connect to something like Discord
or Telegram or something.
Do I want to collect information and send it to somebody?
No thanks.
And it looks like Hermes, the setup has completed.
So let's launch Hermes now.
See if this actually runs.
All right, and now we need to tell it a provider.
You'll pick a provider.
Let's see.
Select a provider and there we go.
And now we are using Claude Opus 4.6,
I guess for the model here.
What is the RAM on this machine?
Let's see if this actually works.
It says Hermes is not connected to any provider yet.
Hermes model allows us to pick one.
Okay, well let's get out of Hermes
and go to Hermes model, run that command.
What do we want to pick here?
All about custom direct API.
There we go.
If I go back into Alaris and go into settings,
you remember this, go to applications.
Let's go to win.
Shared entrance, and copy that particular URL,
and let's see if this will actually take
as the API base URL.
API key, optional, so we'll skip it.
Select API compatibility mode, how about auto-detect.
And then choice one to four, enter to keep detected.
Yeah, detected model, it already detected it, yeah.
Context, length, and tokens.
Leave blank for auto detect.
Display name, it's fine.
Pick whatever you want.
And then select reasoning effort.
Medium is the default.
I'll leave that.
And now we should have connected the Hermes agent
to the Quinn large language model here.
And now what is the RAM of this machine?
And now it's actually, yeah, running the query here.
Let's see, it says, I'll check the RAM on this machine.
It's preparing the terminal, so it opened a terminal
and it ran the command free-h,
a standard Linux terminal command to check the RAM.
I could have done this, I could have opened a terminal.
I just wanted to see, though, if Hermes Agent
was allowed to actually open a terminal,
run that program, and then run the free program
to actually get us the RAM on this machine.
It says this machine has 93 gigabytes of total RAM
with about 58 gigs currently available.
So there you have it. Of course,
Hermes can do a lot more than that along the way,
but now I know it is set up and actually working.
So let me go ahead and close that tab. Hermes,
of course, is still running in the background. I can quit.
Hermes from running in the background if I want,
but that's one I may come back to on
a pretty regular basis, so I'll leave it running.
Now, one thing I do need to mention here
is that as I'm looking at some of the documentation
for the Olaris 1, I do notice that their operating system has had
a recent update that is available
And I just actually updated
the operating system using the Larapass mobile app.
That's how you update the operating system.
I had actually thought I was already
on the latest operating system because if I go into
the marketplace, you do have updates here.
If I click on that, I want to say I updated some things,
but I think I updated just the
applications that were already installed, not necessarily.
The fundamental operating system itself.
But now that I've actually updated the operating system,
when I go into the marketplace, I do have.
Some different apps that are available,
so you have new apps that are available, you want
to make sure you're on the latest version of the operating system,
I believe the latest version of the operating system is 1.12.7,
if I am correct, let me look that up just to make sure, yeah, 1.12.7,
and this latest version comes with some really neat applications such
as Router and Laris, and these are official programs from
Let's click on Router because I believe what this will do,
this will handle some of those connections for us because remember,
each application is its own Kubernetes container.
For example, when we had to do open web UI, you remember we had to go
and connect the Quinn large language model to it.
I think Router allows us to do that in an easier way.
So let's go ahead and click get.
And then click install and that installed correctly.
Let's just go ahead and close that
and go in to the application launcher here
and let's search for our newly installed router program.
And it opens in its own tab and let's go ahead
and take a look at, for example, LLM.
And there is the one large language model
that we've already installed here.
And if I go to the far right hand side of the page
here on the row that has QIN 3.5,
I have this icon right here that says
View Connection Example.
Let me click on that.
And I have this URL here.
Router, dot, and then the rest of the URL.
Let me copy this new router URL.
And what I can use this for is I can now use
this as a way to connect other applications
to the QIN large language model.
For example, OpenWebUI.
If I rerun this, I have the QIN large language model already connected.
Remember, I got the O-Lama URL from the settings menu earlier.
I went to the settings program here,
went to applications, and I went to O-Lama and
shared entrance, and I got this URL.
But now I think I can actually go
to the connections here inside Open Web UI.
I click the connections tab and instead
of manage Ollama API connections, I will actually turn the old URL off.
And now I will add the new URL
and it needs to be an OpenAI API connection.
And let's add URL and I will.
Paste the router URL, and if I click the icon
to verify connection, connection is confirmed there.
And if I save, and then exit out of this,
I still have Quinn here, except now it says
Olaris slash Quinn 3.59b.
So it's connected using the router program.
I think that's neat that we now have this router program
that's gonna manage the connections.
for our various models to the appropriate apps.
Let me close the router program here.
Let me actually go back into the marketplace.
There was one other new application I noticed
now that we've updated the operating system.
There is the Olaris app where you can chat on Olaris
and it's wired to the router program.
So that sounds kind of neat.
Let's go ahead and run an installation.
Of this as well, and as I'm sitting here reading about the Laris app,
it says, of course, it needs an LLM for to be used.
Right. You're chatting with the LLM.
So I've already installed the Quinn large language model.
So that's great. We don't have to install anything else.
And it finished installing.
I could click the open program here from the market out,
but I'll go to the launcher.
Make sure it's here. Yep.
Let's click on Laris and it opens in its own new tab here.
And then we've got our chat box here,
and you can see it's using the QUIN 3.59b model here.
And now I can ask, what system resources am I using?
Just ask it about the machine and see.
If it gives any kind of sensible answer,
it's thinking, it's running a skill, it's running
the program Olaris Dashboard, and then we didn't get an answer.
It says completed in seven seconds,
but it looks like the answer is collapsed for some
reason.
Called tools and commands.
Great, I got the system resources overview from Olaris Dashboard.
Let me present this information in a clear, user-friendly format.
Okay. So I just have to uncollapse all of that.
That's weird that it starts in a folded kind of
format and you have to unfold it,
but it does give me system resources as far as what's available on
the system and what resources I am using,
which programs are using which resources as far as how
much memory and CPU is being used right now by Olamo,
Hermes, Larry's, OpenWebUI, the router
program. So yeah, really cool. Yeah.
Really neat application. So I'm glad I updated the system.
To the latest version of Olaris OS,
which again is version 1.12.7,
you need that new version of the operating system
to have the router program
and to have the Olaris program available for you.
And I also noticed that there were some other applications
that were not here before.
For example, I noticed.
Now that Blender is here,
this was not here earlier when I was searching for it.
And Blender's fantastic.
I've used Blender in the past.
It's for 3D animation modeling,
which I'll be honest, I don't know anything about 3D
animation modeling.
But what I really like about Blender
is it's a pretty decent video editor.
So if I really wanted to, I could use the Alaris One.
for video editing.
And of course, there's plenty of other things
in the marketplace that we actually did not take a look at.
Of course, I can't run through an installation
of all the couple hundred programs here,
but you have some really neat things available.
For example, did you know if you really wanted to,
you could install Windows.
You can run Windows on Aliris if you wanted to.
There is also, I believe, Mac OS is here.
You can run Mac OS on Aliris,
which I think is really,
Pretty friggin' neat that you can run Mac OS on your Alaris One.
So there you have it, a quick overview of the Olaris One AI server.
Now, I mentioned earlier,
we would answer the question,
is this really worth this $5,000 cost, right? Is it worth the
investment to purchase the Olaris One?
I actually think it is. I think this machine, again, it has
some pretty beefy specs. You're getting a lot for that $5,000.
And if you're somebody that's
It's very heavily invested into some of these AI tools.
Now, for me,
I use a lot of AI tools, but I'm not that heavily invested in it.
So, you know, would I purchase one myself?
I don't know if I.
Can justify that cost for me personally,
but many of you guys I know probably can.
If you are somebody, for example,
that is subscribed to a lot of these various AI services,
some of your subscriptions are probably running you north
of $1,000 a year for all the various
things you're subscribed to.
And if you could cut out some of those subscriptions
and host your own AI server locally and run
Run all your own local large language models
and chat assistants and everything.
Could you justify that cost?
I think you probably could.
And that's without even getting
into the whole freedom aspect as far as taking control of
your digital life, right?
And just hosting your own applications without using
these third party services where, you
know, you've got these accounts with these corporations,
with these companies, and you're
really not sure exactly how they're treating your data, for example.
So I think for privacy and security reasons, the Olaris one.
Makes a lot of sense.
Now, what am I gonna use the Olaris One for?
Primarily, I'm gonna use it for large language models
because I'm big into that.
And what I love about the Olaris One
is it's got plenty of storage
for some very large downloads.
Some of these large language models are very big downloads
and I don't have enough storage space,
especially on my home computer.
to run local LLMs. I just, I don't have enough storage, right?
On this machine here at the
office, I do have enough storage, but you know,
the constraints can get kind of tight. If I
install a lot of large language models,
I've got plenty of room though, on the Alaris one for that.
So that's primarily what I will be using it for.
Although I think I'm going to start exploring a
little more of some of these AI image generation tools like comfy UI.
I think that could provide
some interesting content in the future as well.
So I want to thank the Aleris
Thank you to the Oleris team for sending me this unit out to check out.
I also want to thank the patrons of the channel.
You guys, the producers of this video.
Matt, Steve, George, Darla, Flea, Mark,
Methos, Erjan, Peace, Arch, and Fedora,
Realities for Less, Roland, Wargento, and Ubuntu, and Willie.
These guys, they're my highest tiered patrons over on Patreon.
Without these guys, this quick look
at the Oleris one would not have been possible.
The show is also brought to you by each
and every one of these fine ladies and gentlemen.
All these names you're seeing on the screen right now because
I don't have any corporate sponsors.
I'm sponsored by you guys, the community.
If you like my work, want to see more videos about Linux
and free and open source software
like Olaris OS, which runs on the Olaris One,
subscribe to DistroTube over on Patreon.
Peace, guys.