Ollama + OpenWebUI Installation Guide - ANY Model on your PC!

[@66farisfaris]

This was such a helpful guide. It's short, to the point, and easy to follow. When it comes to niche tutorials like this, that's hard to find. Thanks for taking the time to make it!

[@TomsTechAcademy]

Thank you for your kind words, glad you like it!

[@kimberlyallee7582]

This is the only video that has a process that actually works. Thank you so much for this video!

[@TomsTechAcademy]

Thank you for the kind words 😀 glad it helped!

[@ibex33]

Straight to the point, excellent video,thank you.

[@TomsTechAcademy]

Thank you for your comment, glad my video was helpful for you 😀

[@elisabethtovar5949]

Thanks, very useful and straightforward!

[@TomsTechAcademy]

Appreciate your comment, glad my video was useful for you!

[@SaxorTech]

Concise. Thanks!

[@TomsTechAcademy]

You're welcome!

[@asqdekoietozsigarapusha8484]

clean! thank you

[@TomsTechAcademy]

Thank you for your kind words, happy coding 👨‍💻

[@YousofDev]

very good tutorial thank you so much

[@TomsTechAcademy]

Thank you for the kind words, glad my tutorial is useful for you 😀

[@blankslate6393]

Have you manged to run ollama 405B tokens locally?

[@TomsTechAcademy]

Honestly I haven’t tried yet - it requires quite some hardware 😅

[@david.tanasescu]

Nice video! I smell a hint of Dutch. Are you Dutch by any chance?

[@TomsTechAcademy]

Well observed David, I'm in indeed Dutch and recording from the Netherlands :)

[@knobro]

Hi, I am behind a proxy. The standard installation does not work. How do I install the OpenWebUI & model? Thanks

[@ryn1986]

I've seen this installation performed in two different ways. Method 1 involves installing WSL and a Linux distro, then installing Ollama on the Ubuntu instance. Method 2 is installing directly on the Windows PC. Is there a reason to choose one method over the other? Great video, by the way.

[@knobro]

How can I transfer the model to a computer where there is no internet?

[@willowsig9255]

Thank you very much, your video has helped me! But the response to the message is very slow, I have a good computer and a good CG, maybe it's due to my Sata SSD?

[@TomsTechAcademy]

Hi, LLM's will always perform a bit slower on a personal computer than a professional server. I used a machine with 64GB memory and a Geforce RTX3090Ti. If you're stats are comparable, the token generation shouldn't be slow (but still slower than streaming ChatGPT for example). Hope this makes sense :)

[@willowsig9255]

@@TomsTechAcademy Thank you very much for your reply! Yes, I understand that it's faster! I've got 12 gigs on my Nvidia GForce RTX 3060, so that must explain it! I've changed my old SSD for an M.2 SSD, and I find it's a lot faster now! Have a great New Year!

[@midhun3278]

is it possible to run setup in D drive or other than C drive ??

[@ebsmith1102]

so why did you install it before the video? like I am not sure if you understand how an installation guide will work

[@vactum0]

my ollama running same model is deadslow, running in laptop i5 11th gen without GPU 26GB Ram. Is it because of no dedicated GPU?

[@TomsTechAcademy]

It's hard to say why the model is slow on your PC. Don't expect a consumer laptop to provide similar performance as the ChatGPT models (with unlimited hardware) you've probably worked with..

[@willowsig9255]

​@@TomsTechAcademy I'd like to know what configuration you're using in the video. So fast !

[@Furekirou]

thx bro

[@TomsTechAcademy]

Thank you 🙏

[@ArtemioCrovetti-kn7tt]

Thx

[@TomsTechAcademy]

Thank you for your comment :)

[@93simongh]

I have your same setup with ollama installed on my pc and openwebui installed on docker but I can't see anything in models from openwebui, both before and after the ollama run command

[@TomsTechAcademy]

Did you try to refresh OpenWebUI (press F5)? Not sure whether it works but just to be sure :)

[@93simongh]

@@TomsTechAcademy I solved it, had to set network=host

[@TomsTechAcademy]

Thanks for sharing, this could be helpful for others as well 😀

[@nebcat808]

OMG THANK YOU! I've seen COUNTLESS vids on getting this WebUI thing to work, and this literally worked right-off-the-rip with your method! W!

[@TomsTechAcademy]

Thank you for your comment! Glad my video was helpful for you :)

[@SpotAllen]

Dude. That's an SM7B. You should not sound this bad. You're clipping, sir. :)

[@TomsTechAcademy]

Thanks for your comment, I noticed this after uploading this video 😅 in the newer videos this issue has been resolved.