host ALL your AI locally
ฝัง
- เผยแพร่เมื่อ 17 พ.ค. 2024
- Ready to get a job in IT? Start studying RIGHT NOW with ITPro: go.acilearning.com/networkchuck (30% off FOREVER) *affiliate link
Discover how to set up your own powerful, private AI server with NetworkChuck. This step-by-step tutorial covers installing Ollama, deploying a feature-rich web UI, and integrating stable diffusion for image generation. Learn to customize AI models, manage user access, and even add AI capabilities to your note-taking app. Whether you're a tech enthusiast or looking to enhance your workflow, this video provides the knowledge to harness the power of AI on your local machine. Join NetworkChuck on this exciting journey into the world of private AI servers.
📓📓Guide and Commands: ntck.co/ep_401
⌨️⌨️My new keyboard: Keychron Q6 Max: geni.us/0SGY
🖥️🖥️My Computer Build🖥️🖥️
---------------------------------------------------
➡️Lian Li Case: geni.us/B9dtwB7
➡️Motherboard - ASUS X670E-CREATOR PROART WIFI: geni.us/SLonv
➡️CPU - AMD Ryzen 9 7950X3D Raphael AM5 4.2GHz 16-Core: geni.us/UZOZ5
➡️Power Supply - Corsair AX1600i 1600 Watt 80 Plus Titanium: geni.us/O1toG
➡️CPU AIO - Lian Li Galahad II LCD-SL Infinity 360mm Water Cooling Kit: geni.us/uBgF
➡️Storage - Samsung 990 PRO 2TB Samsung: geni.us/hQ5c
➡️RAM - G.Skill Trident Z5 Neo RGB 64GB (2 x 32GB): geni.us/D2sUN
➡️GPU - MSI GeForce RTX 4090 SUPRIM LIQUID X 24G Hybrid Cooling 24GB: geni.us/G5BZ
🔥🔥Join the NetworkChuck Academy!: ntck.co/NCAcademy
**Sponsored by ITProTv from ACI Learning
00:00 - Intro: Building an AI Server for My Daughters
01:01 - Meet Terry: My Powerful AI Server Build
02:02 - Installing PopOS on the AI Server
02:38 - What You Need to Build Your Own Local AI Server
02:58 - Step 1: Installing Ollama AI Foundation
04:24 - Sponsor: ITPro Training from ACI Learning
05:20 - Testing Ollama Installation and Adding Models
06:07 - Interacting with Llama2 AI Model Locally
07:45 - Comparing AI Performance on Different Hardware
08:18 - Step 2: Setting Up Open WebUI
09:31 - Logging into Open WebUI for the First Time
10:06 - Admin Panel: User Management and Permissions
11:07 - Customizing AI Models with Prompts and Restrictions
13:03 - Step 3: Installing Stable Diffusion with Automatic1111
15:52 - Prerequisites and Python Version Management with pyenv
17:21 - Installing Automatic1111 Web UI
18:20 - Testing Stable Diffusion Image Generation Locally
19:23 - Integrating Stable Diffusion into Open Web UI
21:17 - Bonus 1: Using Documents for Context in Open Web UI
21:59 - Bonus 2: Integrating Local AI with Obsidian Note-Taking
23:37 - Wrap-up: The Power and Potential of Local AI
SUPPORT NETWORKCHUCK
---------------------------------------------------
➡️NetworkChuck membership: ntck.co/Premium
☕☕ COFFEE and MERCH: ntck.co/coffee
Check out my new channel: ntck.co/ncclips
🆘🆘NEED HELP?? Join the Discord Server: / discord
STUDY WITH ME on Twitch: bit.ly/nc_twitch
READY TO LEARN??
---------------------------------------------------
-Learn Python: bit.ly/3rzZjzz
-Get your CCNA: bit.ly/nc-ccna
FOLLOW ME EVERYWHERE
---------------------------------------------------
Instagram: / networkchuck
Twitter: / networkchuck
Facebook: / networkchuck
Join the Discord server: bit.ly/nc-discord
AFFILIATES & REFERRALS
---------------------------------------------------
(GEAR I USE...STUFF I RECOMMEND)
My network gear: geni.us/L6wyIUj
Amazon Affiliate Store: www.amazon.com/shop/networkchuck
Buy a Raspberry Pi: geni.us/aBeqAL
Do you want to know how I draw on the screen?? Go to ntck.co/EpicPen and use code NetworkChuck to get 20% off!!
fast and reliable unifi in the cloud: hostifi.com/?via=chuck
1. Set up a private AI server
2. Install Llama for local AI
3. Deploy an AI web UI
4. Integrate stable diffusion AI
5. Generate AI images locally
6. Customize AI models
7. Manage user access for AI server
8. Add AI to note-taking apps
9. NetworkChuck AI server tutorial
10. Run AI on your own hardware
11. Local machine learning server
12. Llama AI installation guide
13. Open source AI server setup
14. Self-hosted AI solutions
15. Enhance workflow with private AI
16. Harness AI power locally
17. AI-powered note-taking
18. Privacy-focused AI tools
19. DIY AI server project
20. Cutting-edge AI technologies
21. Democratizing AI access
22. Empowering users with local AI
23. Offline AI capabilities
24. Future of private AI computing
#aiserver #ollama #llama3 - วิทยาศาสตร์และเทคโนโลยี
Ready to get a job in IT? Start studying RIGHT NOW with ITPro: go.acilearning.com/networkchuck (30% off FOREVER) *affiliate link
Discover how to set up your own powerful, private AI server with NetworkChuck. This step-by-step tutorial covers installing Ollama, deploying a feature-rich web UI, and integrating stable diffusion for image generation. Learn to customize AI models, manage user access, and even add AI capabilities to your note-taking app. Whether you're a tech enthusiast or looking to enhance your workflow, this video provides the knowledge to harness the power of AI on your local machine. Join NetworkChuck on this exciting journey into the world of private AI servers.
📓📓Guide and Commands: ntck.co/ep_401
⌨⌨My new keyboard: Keychron Q6 Max: geni.us/0SGY
🖥🖥My Computer Build🖥🖥
---------------------------------------------------
➡Lian Li Case: geni.us/B9dtwB7
➡Motherboard - ASUS X670E-CREATOR PROART WIFI: geni.us/SLonv
➡CPU - AMD Ryzen 9 7950X3D Raphael AM5 4.2GHz 16-Core: geni.us/UZOZ5
➡Power Supply - Corsair AX1600i 1600 Watt 80 Plus Titanium: geni.us/O1toG
➡CPU AIO - Lian Li Galahad II LCD-SL Infinity 360mm Water Cooling Kit: geni.us/uBgF
➡Storage - Samsung 990 PRO 2TB Samsung: geni.us/hQ5c
➡RAM - G.Skill Trident Z5 Neo RGB 64GB (2 x 32GB): geni.us/D2sUN
➡GPU - MSI GeForce RTX 4090 SUPRIM LIQUID X 24G Hybrid Cooling 24GB: geni.us/G5BZ
🔥🔥Join the NetworkChuck Academy!: ntck.co/NCAcademy
**Sponsored by ITProTv from ACI Learning
first reply
@@MARO_MR Second reply
@@mshark111 third reply
I use chat with rtx. Do you advise me to change to this?
@@MARO_MR LOL
That moment when you realize port 11434 looks like the word llama
Man really gave his kids 2x rtx 4090s for school, he did the "mom i need this [overkill computer] for school"
Its only a $6K build lol
@@brandonwiederhold2573 only $6000 for school...
@@brandonwiederhold2573ONLY 6000? You can adopt me any day
@@notaras1985 exactly
@@notaras1985just do a video for vmware
I'm 62 years old and a computer techy, I'm no super genius though and I'm really happy to have been able to run a local AI on my PC. Private AI is the way to go for sure. I signed up for your free academy for now, there's enough in there to keep me learning/busy for a while yet! :)
Good job pops
now if we can just get some models that have no wokeness/leftist insanity.
@@projectptube I would be happy with an AI that could actually write fairly entry level code instead of churning out garbage code that:
1) won't compile, and efforts to have AI integrated into the development environment correct issues makes it worse with each iteration
2) doesn't actually meet requirements (regardless of how many iterations made to fine tune the output, by which YOU are training the AI)
3) is poorly structured (leading to maintainability problems)
4) lacks proper error handling (leading to problems with stability and data integrity)
5) fails to follow any type of consistent naming convention (code quality/maintainability issues)
6) randomly include variables which determine type on first assignment
7) creates classes where local data types do not correspond to the columns defined in database tables:
7.a) string data types do not enforce the defined length limits
7.b) numeric variables are of inconsistent types
7.c) the data access layer doesn't handle null values, always storing 0 for numeric data types or zero-length strings for (n)varchar fields
8) thrashes database connections (a problem that connection pooling implemented in the client stack doesn't reliably solve)
9) introduces security vulnerabilities.
I could go on, but why bother? The current state of AI for software development is to have companies and sole developers pay to use it while the AI is trained on the well-written source code (or at least better written) the developers end up producing. A packet sniffer will detect that not only is the corrected AI generated code being shared but also proprietary code which has not been authorized for such use.
@@projectptube exactly cough... Gemini... cough. But what do you have in mind when you said that? I am interested to know
This video was an absolute gem, thank you so much. I've been struggling with setting up local AI and the majority of videos I've watched have resulted in me having to try and learn concepts while also deciphering a very heavy accent from the narrator, which made it so much harder for me to focus. This was clear, to the point, and covered everything I wanted. Thank you!
Ollama troubleshooting: if you can’t run Ollama on the first try, open a new terminal and type “Ollama serve”
On my Mac, I had to keep an ollama serve window open and in a new terminal window running the ollama commands would work.
@@ezradevs you do not have to do that to work...
@@Jalan-Api I had to use the ollama serve command on my computer for it to work on WSL, but the windows preveiw works without using the ollama serve command.
Try ollama run llama3
@@nuggetbugget9305 No no, I meant like you do not need the terminal open in background running "ollama serve" on Mac
Man!!! My boss showed me the last local AI video of yours, introducing me to your channel. Now I feel any video you’re making on similar topics I need to see them! Make more videos on this, exploring what all we can do, in workplaces. This is so interesting and cool! Thanks man!
Easy mode: 1. Microcenter's RTX 3090TI x2 (24gb VRAM x2) OR get the Tesla K80's (cheaper) . 2. MOBO that supports either x16 x 2 or x8 x 2. 3. Get at least 64gb system ram (GGUF models run on CPU/RAM/ GPU combined). 4. A 850 - 1,000 Watt power supply. Congrats. You have a computer that almost rivals a system with RTX A6000 (5,000$) card.
Thx Man..
I m building cheap home server for cloud gaming.. for 4 VM : Dell T7810 (200euro) 2x Xeon E5-2697v3 (50euro), ECC 64GB 2400Mhz in quad channel (70euro) Nvidia Tesla P100 16GB (160euro) and added Tesla M40 12G , second PSU 1000w . I hope Llama will use 2 different GPUs. Now the server will be for cloudgaming and AI, so cool :)
tesla k80... dude, your a lifesaver...
i feel seriously dumb for not having found this a year ago...
@@randallrulo2109 something you need to know about the K80's is that it is not a normal PCIe cable needed, it uses a 8 PIN CPU plug. you can get an adapter to convert 2 PCIe 8 pins to 1 8 pin CPU connector
This will greatly help my daughter in the future as we plan to homeschool especially since private GPT can be loaded with local sources like PDF's of books. Very hyped for this content!
Maaaaaan i did this last week on my own, i just had to wait for the master to come along and do it better haha
That’s awesome bro!
Me too!
if you did this alone. be proud of that. don't lessen your achievement. there's enough people out there that will do it as it is. don't help them by doing to yourself.
It all turned out okay. This video helped with Stable Diffusion. Also had some jankyness with WSL networking to work around.
Same 😁
Another fantastic video! And your on screen graphics are some of the best on TH-cam.
I only watched like 4 minutes of your video and I wanted to try asap. Not only did I get it up and running in like an hour but I also configured it to be accessed anywhere in the world I want. Thank you for sparking this fun little piece of technology I can utilize in my own home. This is actually much more useful than I thought because I can have my mother utilize this in her everyday life since I’m all grown up now and out of the house.
can you hint me in a direction for making it accessible from other pcs in a local network?
@@maxhaberstroh2504 Tailscale is probably your easiest solution
+1 in my bucketlist. Sir it’s mindblowing! Combining with vector embedding, local chat history and smart home stuffs will be awesome too
I am using Ollama on my 13 year Old MacBook Pro and it's running pretty fine. Thanks a lot. Keep the great work. Thanks for the videos!! :)!
That is about how old my desktop is. Maybe i have a chance after all.
Good idea ;)
I want to play with this as well. I wound up with a Best Buy open-box i5-12400, 32gb or ram, and an open-box Nvidia 4060 OC 8GB. So I'm in for about $600 all together. I wanted to start as cheap as I could and be power efficient at the same time, at least to start with. Hopefully I'll start playing with it in the next couple of weeks.
One thing I'm curious about though. I wonder how secure these are. Are they really secure, or is it one of those "not too many of them today so nobody is bothering to hack them, yet" situations?
@@Shadow_Banned_Conservative selfhosted LLMs are completly local, there isnt really anything to hack
The magic is that the GPU is more powerful than the average 13yo GPU. In my 15yo pc nothing can run.
Thank you for making it simple! I've followed several tutorials for getting these running locally and they all have their own plus points. Your's with its Stable Diffusion addition is a nice added touch!
Which other videos do you rec?
Dude, your videos are so good. I never miss a video from you. Im working on a project analyzing sports data with local AI for work, so its been very interesting going outside the realm of the simple UIs from OpenAI/Anthropic etc.
People want it local at home, you gotta spend spend spend. I also went crazy and built a terry of my own. My view: having a family AI that you control is very important in the coming future. Also learning the guardrails that you can put on them is very important as well as having some uncensored models when they grow up and are ready for real life. Knowledge is power and these AI models are collections of the vary biggest data sets humanity has ever collected. By the way you helped create these data sets even if you were unaware, thanks big data.
I love these plain simple straight on explanation videos.
A suggestion or addition to this would be:
- how to add or restrict the knowledge base.
For example:
- corporate data, pdf's, tables, pictures, statistics etc and how to purely add this info as knowledge.
- Ask the AI questions and so that it only searches the corporate data and doesn't get blurred with other data.
- let the AI do analysis on the data and pull conclusions on it.
This would be a perfect addition.
No one does it better, NC is awesome. Simple and very intuitive videos.
"- how to add or restrict the knowledge base."
Well, he shows exactly that by showing you the system prompt he gives. You can kinda do whatever you want there, like banning words etc.
Looking into Ollama, you can also train your model on specific data which can help for your your specific uses cases. There is a lot of documentation/videos on that topic on YT if you want.
But that's more relevant of AI training than "easy and fast setup" which was the scope of this video.
check out his last local AI video and his mentions of "Private GPT"
Instead of chatting with models there should be agents with specific skills. why nobody creating something like that?
@@kiranwebros8714 this is what i thought modelfiles were supposed to be, but it doesnt really look like it...
I love this ... I was wondering what PC project I wanted to do next. NOW I KNOW!
Last week, I literally installed Ollama3 but with Python only. Jumping through many hoops to get Cuda recognized. Version had to match Pytorch. Anyway, all works through cli, but love this WebUI install. I'm already playing with local TTS and stoked about Stable Fusion. Thanks for timely video!
Ollama3 🤣
Great video. I've just gone through all of this myself. Looks like they have also added a few more features (LiteLLM, Whisper). Local AI is where it's at!!! Privacy first!!! Hoping they add MemGPT and CrewAI/AutoGen to it.
I was literally thinking of doing exactly this recently, great timing. Thanks.!
This is sweet! Just did this on my spare system and it was faster then I thought it would be.
I9-10900 with 64gb and a SFF Quadro RTX A2000 12gb.
Thank you Chuck
What was faster? These cheap models he is showing? Or got anything better to run?
lol, wish i had a spare system like that! that´s a beast.
Thank you so much! I'm getting ready to setup an AI demo and been wracking my brain on what to do, this is perfect!
"And speaking of...my updates are done." Dude your ad perfectly aligned with my updates as well!
This is some next level content, man!! All love from Brazil
Over the past year, I've incorporated all your tips into my daily routine, and they've definitely helped me feel more energetic and productive! For a while, I even felt unstoppable! But lately, I've been feeling down and sluggish again.
It seems like forcing yourself to do all these things every day can feel like a chore, especially if they're not natural habits. It takes time and effort to build new routines, and it can be frustrating to miss a day. This made me realize that feeling good is more about your mind than your body. When you have a positive outlook, you naturally have more energy.
The key is to find activities that make YOU happy. There's no one-size-fits-all solution! You might find great advice from others, but ultimately, you need to discover what works for you.
I'm not trying to be negative, just to remind everyone that feeling good and bad is a normal part of life. You might have some fantastic days, but there will also be times when you feel down. Don't let that stop you from doing the important things, even if it's just for a few minutes each day. Consistency is key! Successful people might not always feel amazing, but they make time for what matters most in their lives.
Wow! Very impressed by your skills. (Decades ago I was an SCO Sys Admin and watching you handle the console takes me back. Kind of nostalgic.) 🙂
Years ago, I was the SCO admin. I ended up virtualizing the server. It had an ancient application that we needed. I can’t believe I got it to work.
Love the videos you’re making. I’ve been in the IT industry from the late 80’s and still learning everyday. I look forward to your videos as it’s usually something very interesting and has a purpose. I’m already rebuilding an oldish Razer Laptop. Might not have the horsepower but a good test before investing.
Love your videos! Really enjoy watching them *ofc with a cup of warm coffee!* Your videos are so much better "layout" and paced than other guide videos.
That being said, 2 pointers regarding this video.
1st. You would have made Terry ( from nine nine ?) a lot better for AI using a NVIDIA rtx A6000 that cost about the same as your dual GPU combo did.. since AI ( unless configured with accelerate) isnt very good with dual GPU. they cant help each other with the same task. They can however split up the tasks if you configure it. but overall a A6000 would have been better.
2nd. Coming from a hardcore Ubuntu lover.. You gotta check out NixOS.. Its crazy! being able to config the entire OS from 1 file and have different versions of dependencies installed at the same time. its crazy.
Chuck, THIS has got to be the most significant video I've seen in ages. Thank you for sharing this information. I LOVE the idea that we can now have this power under our own control. I will definitely have to do this when I can gather up enough money to build my own Terry (if I'm going to do it I want to do it right).
Money!!! On the first shot too. Epic my friend! Thank you Sir, for keeping it simple.
Thanks, I will be setting this one up for sure - As always, your enthusiasm is awesome..
Thank you for always posting helpful tips. From another IT guy, if you're looking to get into IT or up your game a lot of us out here reference Network Chucks videos for our careers even after we are in the industry. Keep up the magnificent work!
Awesome timing; been tinkering around with Ollama and various models and pondering what my next steps should/could be. Looks like I've got a weekend project!
That just became my new weekend project too haha
Thank you TONS! This is exactly what I needed information-wise to get started!!
Another cool use of local ai integration with other apps I’m using is with VSCode to run your own free local copilot.
Just download extensions like llama coder (text completion) and continue (sidebar chatbot) choose the model and point them to your ollama server and your good to go.
The funny thing is you can see your gpu spiking up when you write code from the code completion suggestions. Welcome to the future! It’s 2024 and now we need a Graphics card to write python scripts xD
This is probably the best video I've seen this year. Please do more AI related content, you're awesome. Already got my CCNA and Azure certs while watching you.
Good tutorial! Never used ollama or open webui. Just gonna say though for anyone trying, loading SD and an LLM at the same time might be a little rough. You'll probably want 32gb of ram and 16gb vram. Might be able to get away with 8gn vram.
Ollama uses llama.cpp as the backend which puts models into the quantized gguf format (allowing it to be split across gpu/cpu vram/ram) for llms.
SD can also load across the gpu/cpu. When you do this for both, the output gets really slow. So you'll want the specs above minimum for both. Also these are 7b llm's which are easier. Terry could run the bigger quantized models which are a lot closer to gpt4 performance.
For everyone else, i'd recommend 7b/8b llama2/3, mistral 7b, and microsoft phi 3b(gguf).
Do you know a model that can do basic math? All the ones i've tried, including the ones you listed, can't answer the simple question "is 24 divisible by 3" correctly. Some will say no, then explain it can be (often in inexplicable ways); others say yes, then no through the explanation, again with odd explanations. (i haven't tried microsoft phi 3b yet though.
@@cakekomo Local models? No. Even opus, gpt4, and Gemini 1.5 still suck at math. There are some math finetuned local models that do slightly better than not, but I wouldn't trust any model for a calculation.
Your best best is gpt4 using python data analysis. If you're looking for math concepts to be explained to you, most LLM's are pretty good since they're trained on math textbooks. Just stick to the calculator or Wolfram for the actual computation.
@@MyWatermelonzdefinitely sticking to pencil and paper or calculators for the actual math, but when somebody is learning and needs a math concept explained to them having AI give a correct answer would be expected.
When one is learning the Divisibilty Rules, and asks an AI if 832,605 is divisible by 3, it attempts to explain the rule and the shows its attempt at the answer. One would expect that answer to be correct.
As I mentioned before, several of the models I tired flubbed it one way or the other. So I simplified the question down from 832,605 during my trouble shooting, and they still flubbed it. One even stated that since “8 + 5 = 3” then 24is divisible by 3 (wtf 😂)!
Thanks for the suggestion!
You are super hyped and surprised about everything, I can't imagine keeping that hype while I know the technicals in such situation. :D
Thank you NetworkChuck! Your videos are both informative and entertaining to watch!
Cheaper alternatives that can be combined with other nvidia GPUs, solely for running AI, are used Nvidia Tesla P40, (24GBof VRAM) currently about ~200 bucks each on the used market. Otherwise go AMD 6800 or newer/better, (16GB+ of VRAM) which are also supported out of the box.
Are you kidding? These go for 7k new. I can see that there are a lot of these offers for used ones, but did you ever confirm that it is legit? Looks like very obvious fraud. Or are you trying to run a scam, yourself?
Would be great if you could make a video on setting up a local AI language model to be trained on documents that get permanently saved in its memory. Seems like there is potential for that using webAI? I want to use this program to be able to reference a part number and have it give me information on the product or manual for that specific part number in my company.
Check out RAG ( retrieval augmented generation). Essentially use a model to store docs into a vector database which is queried by the AI when sending prompts to use in its context window. Lots of videos on RAG out there
This was the best video ive seen in a long time. wow
Subbed.
Just did it! First time I have ever used Linux! Works great! I’ve got llama3 and llava as models. I might bring in llama2 uncensored to see how it goes! This is too cool!
I've had it running - slowly - on a RaspberryPi 5. Love the imploementation on WSL in Windows 11, **BUT** we definitely need a complete guide for those of us who are running an AMD GPU in Windows.
Not everyone had $10K lying around to build a server with TWO $3200CAD Nvidia cards, Chuck...
the updated version of ollama checks amd graphics
@@antonyaustin1388 I found that on the ollama website - unfortunately it looks like the cutoff is 6800XT, right above my 6750XT. Oh well.
I have it running via docker using an old radeon 7 and a ryzen 9 with 12 cores 24 threads and 32gb ram and it runs decently fast on gentoo, and downloaded the auto1111 the way he showed how and its not any slower than his shows.
@@BrandonHurt does it actually use your GPU? If so I'd be interested to see what your docker config is exactly. It runs ok on just my CPU (13700k), but would be faster using the GPU from what I can tell.
i heard your new keyboard immediately, glad you've finally gone custom hotswappable :) very marbly sounding build you got
Yea this is the way to go. More and more homelab stuff is getting easier and easier. Being able to take control of your own data, and run tools like this on your own server is really going to be required.
How much energy is eaten by Terry per month? Do you have any data about this? Real question, I am interested in it.
totally not worth it over regular subscriptions from OpenAi
@@abitw210 I think you haven't watched the video, or you just didn't understand what it is for. He could give a "self prompted" AI for his daughter with limitations. Can you do the same in the OpenAI? And many companies won't share private, sensitive business documents with a third party AI. I can imagine, it is not for you, but it doesn't mean it is not worth it for anybody.
Can we run all this in proxmox
For a "nerdy" video . . . enjoyable to watch, and I appreciate your demonstrations, narrative and humor. Good on you.
OUTSTANDING! And that rig is SICK!
Good luck running anything larger than 8B parameters on just the cpu (and even that might be too big for most people) and expecting more than 2 tokens per second
A relatively recent 8gb gpu is highly recommended to run up to 8B models at over 50 tokens per second
And not just that.. You need to get to something like 100-400B models to be comparable to the bigger AI services.. Those small LLM models are good for things like roleplay and such but when it comes to factual information and productive tasks, they tend to be quite poor.
@@touma-san91 First time i've seen someone mention the comparison to the larger ones. Never knew nor though of that. I might be doing all this work for nothing lol
I run llama3-70B on CPU only I7-13700K and 64gb ddr5. Is it fast, fast? No, but it runs fine.
I can also run it on my 2021 M1 Mac Pro with 64gb of ram. Runs fine there as well.
@@CappellaKeys If you have lot of RAM (Minimum is something like 64 gigs for 70B-models) and good CPU and good GPU with decent chunk of VRAM, you can run these things using GGUF but it will probably take a few minutes to get a response out of the larger models. And you really should use GGUF because that way you can split the load on both the CPU and GPU so it runs tiny bit faster than fully running on CPU.
@@aaroncarroll4158 I'm curious, how fast it is for you? Like how long it takes for it to generate a whole message
Ohoho this is fire! 🔥
I've recently deployed my local AI as well, albeit only with my 6700K (re-using hardware that I already have) and also gave it four 3090s, I think maybe like three weeks ago (whenever Wendell published his video about getting Ollama up and running).
The biggest issue that I've found is that if you download some of the bigger models, it is VERY RAM heavy (both system RAM and VRAM).
Size your model downloads according to the hardware that you have.
The bigger models sometimes can be better, (greater accuracy in the responses), but they also tend to run slower, and again, just eat up more RAM.
LOVING the keyboard sounds, adds a bit of entertaining feedback when doing the compy things.
Everything about your videos are perfect. You are easily in the top 10 of greatest content creators of all time.
Perfect video structure
Perfect sense of humor
Awesome video structure, amazing editing.
You're fast, simple, and concise.
Amazing dude. 👊
As always, a great analysis. Newcomers often wonder if it's too late to navigate the financial market, but the market is always unpredictable. Trading has more advantages than simply holding, so it's important to learn before diving in. Active trades are necessary to ride the market's waves. Thanks to Linda Sue Baier insights, daily trade signals, and my dedication to learning, I've been increasing my daily earnings. Keep it up!
It's unexpected to come across her name here. She understands every beginner’s intention and fix you to a trading course that matches your capacity, she knows her stuff! Her advice has been invaluable to my trading journey. Definitely worth giving a shot!
Investing in alternative income streams that are independent of the government should be the top priority for everyone right now. especially given the global economic crisis we are currently experiencing. Stocks, gold, silver, and virtual currencies are still attractive investments at the moment.
Such market uncertainties are the reason I don’t base my market judgements and decisions on rumors' and here-says, got the best of me 2020 and had me holding worthless position in the market, I had to revamp my entire portfolio through the aid of an advisor, before I started seeing any significant results happens in my portfolio, been using the same advisor and I’ve scaled up 950k within a year, whether a bullish or down market, both makes for good profit, it all depends on where you’re looking.
I’ve been down a ton, I’m only holding on so I can recoup, I really need help, who is this investment-adviser that guides you?
Two 4090's. You really care and love your kids. You are a good father.
Awesome! I was looking for a good starting point and this is a great overview! I'd love to see a follow up video about the Text-to-Speech and Speech-to-Text models that can be integrated with a local AI system.
PS: please support the open source project you use, the devs put in a lot of effort in creating and maintaining them for free, making them accessible for everyone. No pressure tho, enjoy free AI for everyone
3:15 Oh no, a curl piped into a shell… Aargh!
Unjustified panic mode. If you install anything from the internet there is always risk to it no matter the install method. The beauty of an installer script is just you just can read it and make sure it's not doing anything nasty.
@@_modiX
The problem with curl|sh is that a failed download will still get executed. So if the script e.g. had some "rm -rf /tmp/someapp" and the download happened to fail after "rm -rf /", then you can't do anything about it. Or a failed download may cause the partially downloaded script to break and leave you with a broken configuration.
So rather just download the script, quickly check it if it didn't fail (maybe even check the download hash) and _then_ execute it in a seperate step.
Could you describe how to do it your recommended way?
I.E. copy the prompt, but remove " | sh" from the end, and - after SUCCESSFUL download - enter "sh ollama run" ?
@@BruceNJeffAreMyFlies Redirect curl into a file, check the file, and then run it.
@@nikolai00115 Eh, sorry bro. If someone knows how to 'redirect curl into a file, and then run it', they probably already know the answer to my question.
I follow your tutorial and I could install Ollama library and the open-ui in my computer, I have an Rtx 2060 and run very well!
Amazing video!
If you add multiple models for a chat, both models will respond. You will see “< 2 / 2 >” (assuming you added 2 models) beneath the displayed response for you to navigate between the response of each model. The name of the model’s response is displayed above the response. 11:28
Terry seems nice
He has a great personality
I met him in my dream
Have you met Deborah? She is nice to
I will do 0 push-ups for each like.
Perfect comment
@@JUSTREGULARSCREAMINGAAHH perfect reply
21*0 push ups until this comment.
I thought I was doing OK running LLMs on my M3 MBP; Terry is on another level.
Thank you @NetworkChuck! This will be a project for the future. Amazing.
Thank you Lord Jesus for the gift of life and blessings to me and my family $14,120.47 weekly profit Our lord Jesus have lifted up my Life!!!🙏❤️❤️
I'm 37 and have been looking for ways to be successful, please how??
Sure, the investment-advisor that guides me is..
Mrs Kathy lien
Her services is the best, I got a brand new Lambo last week and paid off my mortgage loan thanks to her wonderful services!
Same, I met Kathy lien last year for the first time at a conference in Wilshire, after then my Life has changed for good.God bless Kathy lien
Great walk through, I have been using ollama for a while, love the fact that you have shown how to install stable diffusion,I never expect that all this is possible in my lifetime 75 and still learning , I can't justify building a TERRY , a jetson orin will have to make do,
OMG 1 minute into the video and already loving this content! Keep it going!
Man Terry's a kick-ass gaming rig.
But the disguise as a local AI server is excellent! I'll try and kick this up on my gaming rig as well!
I have been thinking about making a private AI assistant, to help me recall information, set reminders, and basically do alot of work for me.
Maybe a mobile app and browser plugin...etc.
I just dont know where to start but this... This is amazing
@networkchuck, great intro! Would love to see someone going over customizing or training their own model. This would be a great fit for a private hosting platform like yours.
Happy this dropped when i'm about to literally build an AI lab right now.The parts just arrived yesterday :)
That's some serious quality stuff. Brilliant entertaining concept also. I did not know open webui is such a powerhouse app, with all batteries and candies included.
I've been following Brian Roemmele and he is building some really cool models. He doesn't go into the technical details in such an easy way to follow as Network Chuck though so this has been a great help in getting started.
Seriously after 6 months of watching only AI tutorial videos, this is the first time google recommended this epic beard wearing an AI wizard behind it. I have so much work ahead of me and clearly not enough beard in front of me.
Nice home run, NetworkChuck! It's about time.
I did some quick research the other day and came up with roughly the same system parameters for a premium AI server. Two 4090s is already $$$. You really have to want this, or be paid to promote it. You really need this level of power for training models, not running them.
A good project would be to create an AI assistant that mimics a TV AI robot (such as Orac from Blakes 7, or HAL from 2001). You would need a voice to text converter to feed the computer your input, you would need to train the AI to respond like Orac or HAL, and you would need to sample the voice of ORAC or HAL and connect the output to a text to voice converter. We finally then have a real AI assistant like the movies.
PS. I'm running Ollama with a bunch of uncensored models to really explore the possibilities - luv it. Keep up the great work you have earned 1 x subscribe.
Network Chuck is the best! I’m a CCNP, MCSE, CCIA but I always learn new stuff from him. He explains in such a simple and chilled like he know things are just going to work. I’ve found having Linux and coding skill with python and JavaScripy are essential these days. I would love to see a video of a bare metal install of Ubuntu docker container server and run some performance metric comparisons vs a hypervisor and virtual machine deployment. Keep up the great work. You’re an inspiration to us all.
CrewAI is pretty fun to mess around with. Building agents and having them use tools is very awesome.
That's amazing, i already use ollama just as a cli, but using it inside obsidian is even better. Especially since i save sove of the answers in obsidian anyway.
I was able to get this running using a DellT5500 running Ubuntu 24.04, with 65GB RAM, 2-CPU's at 2.13Ghz and a GeForce GTX 1660 Ti . as you can expect it responds slow with the answers BUT it is working! Thank you for the Info video on this topic! this is really amazing to have!! the next step build new PC, wont be as robust as yours but it will be better than T5500...it was an old computer I had Thank you Chuck!!
Fantastic presentation - thank you very much for this. Sent it to a few people - especially for their kids
Hey thanks for all of your contributions!
I’ve been putting a few notes down as I do my usual things and I’ll be happy to put some projects together.
I went to school for industrial electrical stuff and production tech years ago but kept all of my notes and endured many an injury while preparing to go back to school or restart an apprenticeship-so there’s lots of ideas for optimizing task sets and relative workflow!
I’ve been looking at blender and UE5 a bunch over the last few years also, mostly navigating tools & effects and I expect to have something to be able to BLOW YOUR HUMAN MIND with
Not mind blowing yea, but assistance!
ASSISTANCE!
¡ASSISTANCE RUULES!
Chuck i love your vids i just have a hard time with the camera switching alot but i still watch casue its great content
Just love the things you do! Keep it up!
This was a good starting point. I was able to get it running on a vm on an old server. I don't have a gpu in it, but with 20 cores, and a bunch of ram, it is reasonably fast. To be honest, since it is local, it is probably faster that ChatGPT
Your videos are super fun and inspirational❤ and yess please make a video on obsidian it’s my favourite note taking app too
Project suggestion is to work with CrewAI to build the ultimate, local as possible, business management team. Agent-swarm is an awesome upcoming framework but currently only works by utilizing OpenAI. Crew seems to be the best framework for local machine, if its possible.
This is awesome. Well done Chuck and team.
This was absolutely fantastic. I installed on my ASUS laptop which has an integrated GPU. It is pretty snappy, but I want more! I think I build another server with a dedicated GPU. Thanks NetworkChuck.
Bro this is so amazing thanks for all you doing
Great video and very informative Chuck! I had to chuckle when you called Automatic1111 "Automatic one one one one", it's actually "Automatic eleven eleven"
Thanks for sharing this! You've inspired me to start my local LLM chatbot for my fam too.
Love the video, but I even more love your streaming set up. Do you have any videos on what you all use for screen casting along with a cool zoom in out and face bubble?
Dear NetworkChuck, your clips are amazing. Just a quick note: when I installed ollama on my windows 10, at 6:35 the ollama was not already running after installling it with your curl command into Ubuntu from Windows Store so I had to launch it manually with "ollama serve" in another ubuntu window... However it was running when installing the same way on windows 11. Thanks for sharing your findings, I like the idea of all of us holding hands and sing... LOL :)
I also am glad you are sponsored, because 2x 4090's are really out of my budget so I used 2x 3060ti's and its not as fast as yours but it works
Yeah, most of us have to live in the real world