Well, first, it was more for me.

首先，這更多是為了我自己。

I wanted to run all of my AI locally.

我想在在地運行所有人工智能。

And I'm not just talking command line with Olama.

我說的不僅僅是使用 Olama 的命令行。

No, no, no.

不，不，不

We have a GUI, a beautiful chat interface.

我們有一個圖形用戶界面，一個漂亮的哈拉界面。

And this thing's feature-filled.

這東西功能齊全。

It's got RBAC, chat histories, multiple models.

它有 RBAC、聊天記錄和多種模型。

We can even add stable diffusion.

我們甚至可以增加穩定的擴散。

And I was able to add this to my notes application, Obsidian, and have my chat interface right there.

我可以將其添加到我的筆記應用程序 Obsidian 中，並在那裡設置哈拉界面。

I'm gonna show you how to do this.

我來教你怎麼做。

Now, you don't need something crazy like Terry.

現在，你不需要像特里那樣瘋狂。

That's what I named my AI server.

這就是我給人工智能服務器起的名字。

It can be something as simple as this, this laptop.

就像這檯筆記本電腦一樣簡單。

I'll actually demo the entire setup on this laptop.

實際上，我會在這檯筆記本電腦上演示整個設置。

So luckily, the computer you're using right now, the one you're watching this video on, will probably work.

所以幸運的是，你現在使用的電腦，也就是你正在觀看這段視頻的電腦，很可能可以正常工作。

And seriously, you're gonna love this.

說真的，你一定會喜歡的。

It's customizable.

可定製。

It's wicked fast, like way faster than anything else I've used.

它的速度非常快，比我用過的任何其他產品都要快。

Isn't that amazing?

是不是很神奇？

And again, it's local.

同樣，它也是在地的。

It's private.

它是私人的。

I control it, which is important because I'm giving it to my daughters.

我控制著它，這一點很重要，因為我要把它交給我的女兒們。

I want them to be able to use AI to help with school, but I don't want them to cheat or do anything else weird.

我希望他們能利用人工智能幫助學習，但我不希望他們作弊或做其他奇怪的事情。

But because I have control,

但因為我有控制權、

I can put in special model files that restrict what they can do, what they can ask.

我可以在特殊模型文件中限制他們的行為和要求。

And I'll show you how to do that.

我會告訴你如何做到這一點。

So here we go.

那麼，我們開始吧。

We're about to dive in.

我們即將潛入水中。

But first, let me have you meet Terry.

但首先，讓我給你介紹一下特瑞。

Now, Terry has a lot of muscle.

現在，特里肌肉發達。

So for the case, I needed something big.

是以，我需要一個大箱子。

I got the Leon Lee 011 Dynamic Evo XL.

我買的是 Leon Lee 011 Dynamic Evo XL。

It's a full tower E-ATX case.

這是一款全塔式 E-ATX 機箱。

Perfect to hold my Asus X670-E Creator ProArt motherboard.

非常適合放置我的華碩 X670-E Creator ProArt 主板。

This thing's also a beast.

這東西也是一頭猛獸。

I'll put it in the description so you can look at it.

我會把它寫在描述裡，這樣你就能看到了。

Now, I also gave Terry a big brain.

現在，我還給了特瑞一個大腦袋。

He's got the AMD Ryzen 9 7950X.

他使用的是 AMD Ryzen 9 7950X。

That's 4.2 gigahertz and 16 cores.

這相當於 4.2 千兆赫和 16 個內核。

For memory, I went a little crazy.

為了記憶，我有點瘋狂。

I've got 128 gigabytes of the G.Skill Trident Z5 Neo.

我有 128GB 的 G.Skill Trident Z5 Neo。

It's DDR5-6000 and way overkill for what I'm doing.

它是 DDR5-6000，對我的工作來說太誇張了。

I think.

我想是的

I got a Leon Lee water cooler for the CPU.

我為 CPU 安裝了 Leon Lee 水冷卻器。

I'm not sure if I'm saying Leon Lee right.

我不知道我這樣說李昂對不對。

I don't know.

我不知道。

Correct me in the comments.

請在評論中指正我。

You always do.

你總是這樣。

And then for the stuff AI loves, I got two 4090s.

為了滿足人工智能的需求，我買了兩臺 4090。

It's the MSI Supreme and they're liquid cooled so they could fit on my motherboard.

這是微星至尊版，它們是液冷的，所以可以安裝在我的主板上。

24 gigabytes of memory each, giving me plenty of muscle for my AI models.

每臺機器的內存為 24 千兆字節，為我的人工智能模型提供了充足的動力。

For storage, we got two Samsung 990 Pros, two terabytes, which you can't see because they're behind stuff.

在存儲方面，我們有兩臺三星 990 Pro，兩 TB，你看不到它們，因為它們在東西后面。

And also a Corsair AX1600i power supply.

還有一個海盜船 AX1600i 電源。

1600 watts to power the entire build.

1600 瓦的功率可為整個構建提供動力。

Terry is ready.

特里準備好了

Now I'm surprised to say my system actually posted on the first attempt, which is amazing.

現在，我很驚訝地發現，我的系統居然在第一次嘗試時就發佈了，這真是太神奇了。

But what's not amazing is the fact that Ubuntu would not install.

但令人驚訝的是，Ubuntu 無法安裝。

I tried for hours, actually for a whole day.

我試了好幾個小時，實際上試了一整天。

And I almost gave up and installed Windows, but I said, no, Chuck, you're installing Linux.

我差點放棄安裝 Windows，但我說，不，查克，你要安裝 Linux。

So I tried something new, something I've never messed with before.

於是，我嘗試了一些新的東西，一些我以前從未嘗試過的東西。

It's called Pop OS by System76.

這就是 System76 的 Pop OS。

This thing is awesome.

這東西太棒了。

It worked the first time.

第一次就成功了。

It even had a special image with NVIDIA drivers built in.

它甚至還有一個內置英偉達驅動程序的特殊鏡像。

It just stinking worked.

它就是臭氣熏天地工作著。

So I sipped some coffee, didn't question the magic and moved on.

於是，我喝了幾口咖啡，沒有質疑魔法，繼續前行。

Now, if you do want to build something similar,

現在，如果你確實想建造類似的東西、

I've got all the links below.

下面是所有鏈接。

But anyways, let's talk about how to build your very own local AI server.

無論如何，讓我們來談談如何建立自己的在地人工智能服務器。

First, what do you need?

首先，您需要什麼？

Really, all you'll need is a computer.

真的，你只需要一臺電腦。

That's it.

就是這樣。

It can be any computer running Windows, Mac or Linux.

它可以是任何運行 Windows、Mac 或 Linux 的電腦。

And if you have a GPU, you'll have a much better time.

如果你有 GPU，你會玩得更開心。

Now, again, I have to emphasize this.

現在，我必須再次強調這一點。

You won't need something as beefy as Terry, but the more powerful your computer is, the better time you'll have.

您不需要像特里那樣強大的電腦，但您的電腦越強大，您的時間就越充裕。

Don't come at me with a Chromebook, please.

請不要用 Chromebook 來對付我。

Now, step one, Olama.

現在，第一步，奧拉瑪。

This is the foundation for all of our AI stuff and what we'll use to run AI models.

這是我們所有人工智能產品的基礎，也是我們用來運行人工智能模型的基礎。

So we'll head on over to olama.ai and click on download.

是以，我們將前往 olama.ai，點擊下載。

And they get a flavor for every OS.

每個作業系統都有自己的口味。

I love that.

我喜歡這樣。

Now, if you're on Mac, just download it right now and run it.

如果你用的是 Mac，現在就下載並運行它。

If you're on Windows, they do have a preview version, but I don't want you to do that.

如果你使用的是 Windows 系統，他們確實有一個預覽版，但我不希望你這樣做。

Instead, I want you to try the Linux version.

相反，我想讓你試試 Linux 版本。

We can install it with one command.

我們只需一條命令就能安裝它。

And yes, you can run Linux on Windows with WSL.

是的，你可以使用 WSL 在 Windows 上運行 Linux。

Let's get that going real quick.

讓我們快點開始吧。

First thing I'll do is go to the start bar and search for terminal.

我要做的第一件事就是在開始欄中搜索終端。

I launched my terminal.

我啟動了我的終端。

Now, this first bit is for Windows folks only,

現在，第一點僅供 Windows 用戶使用、

Linux people, just hang on for a moment.

Linux 用戶，請稍等片刻。

We got to get WSL installed or the Windows subsystem for Linux.

我們必須安裝 WSL 或 Linux 的 Windows 子系統。

It's only one command, WSL dash dash install.

只有一個命令：WSL dash dash install。

And that's it actually.

實際上就是這樣。

Hit enter, and it's gonna start doing some stuff.

按下回車鍵，它就會開始做一些事情。

When it's done, we'll set up a username and password.

完成後，我們將設置用戶名和密碼。

I got a new keyboard, by the way.

順便說一句，我買了一個新鍵盤。

Do you hear that?

你聽到了嗎？

Link below, it's my favorite keyboard in the entire world.

鏈接如下，這是我在全世界最喜歡的鍵盤。

Now, some of you may have to reboot, that's fine.

現在，有些人可能需要重新啟動，沒關係。

Just pause the video and come back.

暫停視頻，然後再回來。

Mine is ready to go though, and we're rocking Ubuntu 22.04, which is still amazing to me that we're running Linux on Windows.

不過，我的設備已經準備就緒，我們正在使用 Ubuntu 22.04，在 Windows 上運行 Linux 仍然讓我感到不可思議。

That's just magic, right?

這就是魔法，對嗎？

Now, we're about to install Olama, but before we do that, you got to do some best practice stuff, like updating our packages.

現在，我們要安裝 Olama，但在此之前，你得做一些最佳實踐，比如更新我們的套裝軟體。

So we'll do a sudo apt update, and then we'll do a sudo apt upgrade dash y to apply all those updates.

是以，我們將執行 sudo apt update，然後執行 sudo apt upgrade dash y 來應用所有這些更新。

And actually, while it's updating, can I tell you something about our sponsor?

實際上，在更新的同時，我能告訴你一些關於我們贊助商的事情嗎？

IT Pro by ACI Learning.

ACI Learning 的 IT Pro。

Now, in this video, we're gonna be doing lots of heavy Linux things.

現在，在這段視頻中，我們要做很多 Linux 的重頭戲。

I'm gonna walk you through it.

我陪你走過去

I'm gonna hold your hand, and you may not really understand what's happening.

我會握著你的手，你可能真的不明白髮生了什麼。

That's where IT Pro comes in.

這就是 IT Pro 的作用所在。

If you want to learn Linux or really anything in IT, they are your go-to.

如果你想學習 Linux 或 IT 方面的任何知識，他們都是你的首選。

That's what I use to learn new stuff.

我就是用它來學習新知識的。

So if you want to learn Linux to get better at this stuff, or you want to start making this whole hobby thing your career, actually learn some skills, get some certifications, get your A+, get your CCNA, get your AWS certifications, your Azure certifications, and go down this crazy IT path, which is incredible, and it's the whole reason I make this channel and make these videos.

是以，如果你想通過學習 Linux 來更好地掌握這些東西，或者你想把這個愛好變成你的職業，那就學習一些技能，獲得一些認證，獲得 A+、CCNA、AWS 認證、Azure 認證，沿著這條瘋狂的 IT 之路走下去，這太不可思議了，這也是我製作這個頻道和這些視頻的全部原因。

Check out IT Pro.

查看 IT Pro。

They've got IT training that won't put you to sleep.

他們提供的 IT 培訓不會讓你昏昏欲睡。

They have labs, they have practice exams, and if you use my code NetworkChuck right now, you'll get 30% off forever.

他們有實驗室，有模擬考試，如果你現在使用我的代碼 NetworkChuck，你將永遠享受 30% 的折扣。

So go learn some Linux, and thank you to IT Pro for sponsoring this video and making things like this possible.

所以，快去學習 Linux 吧！感謝 IT Pro 贊助本視頻，讓這樣的事情成為可能。

And speaking of, my updates are done.

說到這裡，我的更新已經完成了。

And by the way, I will have a guide for this entire thing.

順便說一句，我會為整個活動準備一份指南。

Every step, all the commands, you can find it at the free NetworkChuck Academy membership.

每個步驟、所有命令，您都可以在免費的網絡查克學院會員資格中找到。

Click the link below to join and get some other cool stuff as well.

點擊下面的鏈接加入，還能獲得其他很酷的東西。

I can't wait to see you there.

我迫不及待地想在那裡見到你。

Now we can install Olama with one command.

現在，我們只需一條命令就能安裝 Olama。

And again, all commands are below.

同樣，所有命令都在下面。

Just gonna paste this in.

我只是想把這個粘貼進去。

A nice little curl command, little magic stuff, and this, I love how easy this is, watch.

一個漂亮的小卷發指令，一個小魔術，還有這個，我喜歡這個，看，多麼簡單。

You just sit there and let it happen.

你只需坐在那裡，順其自然。

Do you not feel like a wizard when you're installing stuff like this?

安裝這樣的東西時，你不覺得自己像個魔法師嗎？

And the fact that you're installing AI right now, come on.

而且你現在就在安裝人工智能，得了吧。

Now notice one thing real quick.

現在請迅速注意一件事。

Olama did automatically find out that I have an NVIDIA GPU, and it's like, awesome, you're gonna have a great time.

Olama 自動發現我有一個英偉達™（NVIDIA®）GPU，然後就說，太棒了，你會玩得很開心的。

If it didn't see that and you do have a GPU, you may have to install some NVIDIA CUDA drivers.

如果它沒有看到，而你確實有 GPU，你可能需要安裝一些英偉達™（NVIDIA®）CUDA 驅動程序。

I'll put a link for that below, but not everyone will have to do that.

我將在下面提供一個鏈接，但不是每個人都必須這樣做。

And if you're rocking a Mac with an M1 through M3 chip, you're gonna have a good time too.

如果你使用的是裝有 M1 至 M3 芯片的 Mac，你也會玩得很開心。

They will use the embedded GPU.

它們將使用嵌入式 GPU。

Now at this point, our Mac users, our Linux users, and our Windows users are all converged.

在這一點上，我們的 Mac 用戶、Linux 用戶和 Windows 用戶都趨於一致。

We're on the same path.

我們走的是同一條路。

Welcome, we can hold hands and sing.

歡迎，我們可以手拉手一起唱歌。

It's getting weird.

越來越奇怪了

Anyways, first we have to test a few things to make sure Olama is working.

總之，我們首先要測試一些東西，以確保 Olama 能正常工作。

And for that, we're gonna open our web browser.

為此，我們要打開網絡瀏覽器。

I know, it's kind of weird.

我知道，這有點奇怪。

Just stick with me.

堅持住

I'm gonna launch Chrome here, and here in my address bar, I'm gonna type in localhost, which is looking right here at my computer, and port 11434.

我要啟動 Chrome 瀏覽器，在地址欄中輸入 localhost，也就是我的電腦，然後輸入 11434 端口。

Hit enter, and if you see this right here, this message, you're good to go, and you're about to find this out.

點擊回車鍵，如果你在這裡看到這條資訊，你就可以開始了，你馬上就會發現這一點。

Port 11434 is what Olama's API service is running on, and it's how our other stuff is gonna interact with it.

11434 端口是 Olama API 服務的運行端口，我們的其他程序也將通過它進行交互。

It's so powerful, just check this out.

它太強大了，看看這個。

I'm so excited to show you this.

我很高興能給你看這個。

Now, before we move on, let's go ahead and add an AI model to Olama.

現在，在繼續之前，我們先為 Olama 添加一個人工智能模型。

And we can do that right now with olama pull, and we'll pull down Lama 2, a very popular one.

我們現在就可以通過 olama pull 來做到這一點，我們將拉下 Lama 2，這是一款非常受歡迎的產品。

Hit enter, and it's ready.

點擊回車，就可以了。

Now, let's test it out real quick.

現在，讓我們來快速測試一下。

We'll do olama run Lama 2.

我們將進行 "喇嘛跑 2"。

And if this is your first time doing this, this is kind of magic.

如果這是你第一次這樣做，這就有點神奇了。

We're about to interact with a chat GPT-like AI right here.

我們將在這裡與一個類似 GPT 的人工智能哈拉互動。

No internet required, it's all just happening in that five gigabyte file.

無需聯網，一切都在這個五千兆字節的文件中發生。

Tell me about the solar eclipse.

跟我說說日食吧

Boom, and you can actually control C that to stop it.

嘭，你其實可以控制 C 來阻止它。

Now, I wanna show you this.

現在，我想給你看這個

I'm gonna open up a new window.

我要打開一個新窗口。

This is actually an awesome command.

這其實是一個很棒的命令。

And with this WSL command,

使用 WSL 命令

I'm just connecting to the same instance again, a new window.

我只是再次連接到同一個實例，一個新窗口。

I'm gonna type in watch-n 0.5, not four, five, nvidia-smi.

我要輸入 watch-n 0.5，而不是 4、5、nvidia-smi。

This is going to watch the performance of my GPU right here in the terminal and keep refreshing.

它將在終端上觀察我的 GPU 性能，並不斷刷新。

So keep an eye on this right here as I chat with Lama 2.

請繼續關注我與喇嘛 2 號的對話。

Lama 2, give me a list of all Adam Sandler movies.

喇嘛 2，給我一份亞當-桑德勒所有電影的清單。

And look at that GPU go, ah, it's so fun.

再看看 GPU，啊，太有趣了。

Now, can I show you what Terry does real quick?

現在，我可以讓你看看特里是怎麼做的嗎？

I gotta show you Terry.

我得讓你看看特瑞

Terry has two GPUs.

特里有兩個 GPU。

Here they are right here.

它們就在這裡。

And olama can actually use both of them at the same time.

實際上，Olama 可以同時使用這兩個功能。

Check this out, it's so cool.

看看這個，太酷了。

List all the Samuel L. Jackson movies.

列出塞繆爾-傑克遜的所有電影。

And look at that.

看看這個。

Isn't that amazing?

是不是很神奇？

And look how fast it went.

看看它走得多快。

That's ridiculous.

這太荒謬了。

This is just the beginning.

這僅僅是個開始。

So anyways, I had to show you Terry.

總之，我得讓你看看特里。

So now we have olama installed.

現在我們已經安裝了 olama。

That's just our base.

這只是我們的基地。

Remember, I'm gonna say bye.

記住，我要說再見了

So forward slash bye to end that session.

是以，向前斜線 "拜拜 "結束了這次會話。

Step two is all about the web UI.

第二步是網絡用戶界面。

And this thing is amazing.

這東西太神奇了

It's called OpenWebUI.

它叫 OpenWebUI。

And it's actually one of many web UIs you can get for olama.

實際上，它是 olama 的眾多網絡用戶界面之一。

But I think OpenWebUI is the best.

但我認為 OpenWebUI 是最好的。

Now OpenWebUI will be run inside a Docker container.

現在，OpenWebUI 將在 Docker 容器中運行。

So you will need Docker installed.

是以，你需要安裝 Docker。

And we'll do that right now.

我們現在就去做。

So we'll just copy and paste the commands from Network Shrug Academy.

是以，我們只需複製並粘貼網絡聳肩學院的命令即可。

This is also available on Docker's website.

這也可以在 Docker 網站上找到。

First step is updating our repositories and getting Docker's GPG key.

第一步是更新我們的軟件源並獲取 Docker 的 GPG 密鑰。

And then with one command, we'll install Docker and all its goodies.

然後只需一條命令，我們就能安裝 Docker 及其所有好東西。

Ready, set, go.

準備，準備，開始

Yes, let's do it.

是的，我們開始吧。

And now with Docker installed, we'll use it to deploy our OpenWebUI container.

現在，Docker 已安裝完畢，我們將用它來部署 OpenWebUI 容器。

There'll be one command.

只有一個命令

You can simply copy and paste this.

您只需複製並粘貼即可。

This docker run command is going to pull this image to run this container from OpenWebUI.

這個 docker run 命令將從 OpenWebUI 中調用這個鏡像來運行這個容器。

It's looking at your local computer for the olama base URL, because it's going to integrate and use olama.

它會在在地計算機上查找 olama 基本 URL，因為它會集成並使用 olama。

And it's going to be using the host network adapter to make things nice and easy.

它將使用主機網絡適配器，使事情變得簡單易行。

Keeping in mind, this will use port 8080 on whatever system you're using.

請注意，無論您使用什麼系統，都將使用 8080 端口。

Now all we have to do is hit enter after we add some sudo at the beginning, sudo docker run, and let it do its thing.