Installing Open WebUI Ollama Local Chat with LLMs and Documents without Docker

[@marcocinalli755]

thank you! it seems much lighter than uising docker... how can i update it? through git pull?

[@patrickdali4063]

thank you so much, this tutorial was extremely clear and useful ! you won a sub !

[@MarkAnthonyfernandez-j9h]

i happen to installed ollama on docker and i wan to try this can I clean up my existing ollama then follow this method?

[@luisfernandogarciagonzales9566]

Amazing, thanks From Honduras.

[@esmail88]

perfect , clear to the point , funny

[@vsurp]

'open-webui' is not recognized as an internal or external command, operable program or batch file. :/

[@Natlamir]

Sounds like something might have gone wrong during the pip install open-webui?

[@vsurp]

@@Natlamir i honestly don't know, i've tried setting different paths for installation of python and all that.

[@youtubeccia9276]

best instruction guy on yt :D

[@Natlamir]

Thank you for the kind words!

[@RKFlago]

Is it possible to install open-webui & ollama in another drive (like d drive)

[@kidsstoriezhub]

bhai ye narrator kaha se lia, thats more interesting than the entire process xD

[@NoobMad5551]

I got a problem when command open-webui serve

[@MarcoAWilcox619]

Thank you! Pretty efficient way to save memory. In my case saved 5GB compared to running open webui with docker.

[@Natlamir]

That is great to hear, saving you 5GB! Thanks for sharing your experience.

[@Tech-era]

How do I run the model offline? once I disable the internet, the whole process just stops

[@Natlamir]

Is that so? Interesting. I will try that. Does that happen for you when using just ollama through the command prompt, or only when using it with WebUI?

[@Tech-era]

@@Natlamir only happens when using llama on webui. It however works fine on the command prompt whether in offline or online mode

[@Natlamir]

@@Tech-era Thanks for the info, I will try that out and will see if the same happens for me when trying through webui.

[@Natlamir]

@@Tech-era I think I found a solution. You will need to add an environment variable. A user variable. Add this: Variable name: OLLAMA_BASE_URLS Variable value: 127.0.0.1:11434 then relaunch the conda prompt and try again.

[@brug6998]

how can i update it without uninstalling then reinstalling?

[@Natlamir]

Good question, you should be able to use the upgrade flag when you want to update it. something like this: pip install --upgrade open-webui

[@brug6998]

@@Natlamir Thanks so much!

[@kingofmayy]

Hey can you make a video next about how to integrate Fabric GitHub project into the mix, would be super cool if i could use custom prompts inside webui.❤

[@Natlamir]

Thanks for the suggestion! That is a great idea, I will look into integrating fabric patterns for different types of prompts.

[@kingofmayy]

@@Natlamir I can't wait to see the vid, trying to figure it out for myself as well.

[@DarrenAddy]

I second the Fabric integration!

[@altf4urlife]

If I use this method, will it run on the CPU or GPU?

[@Natlamir]

Should run on GPU. When I run it, it is utilizing GPU / CUDA

[@JoshVonhauger]

Great video i am new to conda. How do you SSL cert the environment

[@Natlamir]

Thanks! I am not too familiar with that, but they seem to have some detail in the Hosting section: docs.openwebui.com/tutorial/apache/

[@receps.8396]

what is the name of the narrator? Man?

[@codingchannel6263]

that stpis voice AI is annoying