วีดีโอ

CPU: Minimum: Modern processor with at least 4 cores. RAM: 7B Models: At least 8 GB. 13B Models: At least 16 GB. NVIDIA GPUs: Compute capability of at least 5.0. AMD GPUs: Supported for enhanced performance VRAM Requirements: 7B Models: 8 GB VRAM.

Hi vicky, you need to download a model. It’s under the admin settings and models section.

Thanks Louis!

Clude Ai

@mesferalwaked3 do thay have an open source model?

Hey! CodeGemma 7b is a pretty good coding model : ollama.com/library/codegemma

The best small model right now is Qwen2.5-Coder by far. It is a 7b model that is a beast, it run anywhere at just 7b and is fast. There is a 33b version of it in the way, I hope that it comes just as awesome. If you want something bigger in local I don't know, I just go with Qwen2.5-Coder and if need something better I go with Claude, Qwen us so go that it is not even worthyfor me going with others at 30b and 70b is to big for my laptop.

@@lelouchlamperouge5910 thank you very much, just checked it out. What about the best image to code LLM? Do you know any?

Thanks Maksyudin!

I believe it was the same prompt. I was experimenting with different models. Let me double-check!

Thank you!

Thanks!

Thank you, your support truly means a lot! :)

Hi! I’m updating this video, new one will be out by tomorrow! Thanks for letting me know!

Hello @nullnvoid7, the new video is out! th-cam.com/video/lqKapMX2GAI/w-d-xo.html

Hi! I’m updating this video, new one will be out by tomorrow! Thanks for letting me know!

Hi @christiand2426, I have updated the video to the new version of OpenWebUI here - th-cam.com/video/lqKapMX2GAI/w-d-xo.html

Hey @jakkalsvibes, You'd click on the "Workspace" first, and that'll take you to where you can access the Documents and Prompt tabs. Let me know if it works!

It was not great for me at first as well, but if you change some of those admin settings and experiment with different models, the model results are quite decent.

Good point Shashikant! Always be diligent in using functions and only use the ones vetted by OpenWebUi

Hi Serg yes I have been using Groq to do this for months and I haven't had to pay anything yet.

@@DigitalBrainBase wow, thanks a lot

Hey serg, unfortunately the payment through OpenAI is directly through their website as far as I know. I'll let someone else comment if they know, but I don't think it's possible through OpenWebUI.

Hi @fredkzk, To be honest, plain text files are going to be the simplest and fastest to process because they are just raw text without any formatting. My first preference would always be .txt files.

Hey, good call! Let me update the description ASAP!

Hello Rocketbox9, Eleven has a bunch of free voices available. You could also look at voice-vector or Unreal Speech.

Hi @pleabargain, Good call! I’ll be talking about that in a future video.

Hey Fred, Good observation! I only wanted two for the demo. I didn’t enable more.

Hey Fred, Looks like we both taught each other something new! How did you like the Eleven Lab API?

@@DigitalBrainBase Haven't tried it yet. Considering it for AI coding assistance tutorial...

Thanks for the comment! If you have any questions or would suggestions for material you would like us to cover please let us know!

@@DigitalBrainBase I'm studying the deep use of functions and pipelines features for setting up a team of dynamically created agents. So any material abt that challenging feature would be welcome.

google lens can already do this for free why would you pay for chatgpt4o to do this

@@genericuser1454 correct!

@@Daniel_Lah Sure thing Daniel thanks for watching! Dave

Thanks for the comment Andre! The benefit is that on ChatGPT you can only use Open AI’s models and all the history, memory, etc generated when you use it is locked within chatgpt. When using Open WebUI you can still use Open AI’s models, but you can also use any of the many other models that are available including open source models that you can run 100% privately and securely on you computer without even having to g to be connected to the internet. So with Open WebUI you own the retrieval and storage/memory of your digital brain and can use any intelligence you want with it. with open ai they own it all and you can only use their intelligence. Does that answer your question? Thanks again Dave

@@DigitalBrainBase Thank you for your response! I understand the benefits of using the other models and being able to use it even offline, but as you also used an open ai model in the video I got curious if integrated with OpenWebui it would somehow be different from doing other ways to call open ai api directly. I guess it doesn't! Nevertheless, Open Web Ui is exactly what was looking for, thanks for this great tutorial!

@@andrehenriques2163 cool glad you liked it and yes nothing different that I can think of right now but I bet we discover some cool differences as we continue to dig in. Will come back here and reply if/when we do. Thanks again for the comment and the feedback.

Thanks for the comment! Beck is out on vacation but will discuss with him and revert when he gets back.

@@DigitalBrainBase I tried editing the code and all… I had some issues hosting it… can you guys do a tutorial on hosting and editing the code . Many people interested I am sure they will watch.

@@ACTFIREE Yes we are planning to do both looking for someone to help me create now.

@@DigitalBrainBase amazing can’t wait to watch

Ok Beck is back from vacation and I discussed with him this morning. We host everything on one virtual machine so this is not an issue. Apparently however Open WebUI just did a big update that may allow us to separate the front and backend and maintain security so we are looking into that now. Hope that helps! Dave

Thanks for the Idea Ricky!

Hi Ricky thanks for the comment. I am not familiar with Proton Server specifically but you should be able to host Open WebUI wherever you can host a website. My developer set it up on Google Cloud and here are the steps that he followed: docs.google.com/document/d/1Iu1PYZ2gagN8boVXviZm_YDoVcw9heKdaZaY5gaCk9c/edit?usp=sharing Hope that helps let me know if you have any other questions or comments!

Thanks for watching and for the comment! My goal is to continue moving up the complexity curve in my own use of Open WebUI and sharing my learnings as I go. All of the areas you mentioned are on my list so stay tuned for more on those soon! 😃

@@jim02377 Thanks Jim!