Відео

What is the function of image regularization, then what kind of regularization image should be chosen to create a facial lora? thank you

Wow you explained what at least 15 other channels failed to do, THANKS SO MUCH! I was so confused :) :) :)

Well explained, but I am stuck on 'zsh: no such file or directory: ./webui.sh'. I do not know what to do. Any suggestion would be highly appreciated

Is this video outdated and the command brew install cmake protobuf rust python@3.10 git wget doesn't work anymore? Because every time I do copy paste, is says "command not found". What should I do? Thank you

She knows her stuff but I'm slow. I just needed: Here's a Lora, here's step by step to use it with SD. I'm looking for another vid that does that. LOL

Does it works with base pony chackpoints and loras?

love how she subtedly moves past the hentai and porn categories lmfao

flux for forge! check it out

So.. I do need to have each model separate for generating a preprocessed input for the Union model, correct?

Thanks

I was curious how ADetailer would work on a photo of my fat cat & it was hilarious 😂😂😂 for some reason ADetailer gave my cat human eyes & hands 😂😂😂😂 but mainly user error too

Automatic 1111 has feature to adjust mask preview and see changes of mask after adjusting values (forget name), but can you do that in comfy?

thank you for your videos and your sexy voice :) im trying this week to to learn how to try lora`s. The main kohya gave me errors on installs, so trying the portable 1 click install and have that installed. I have 2 gpus, 1 8gb and external 12gb. how do i tell it what one to use please.

flux guide?

Does this still work for you today? Mine started throwing errors over the weekend.

Thanks

Can you Pleeeeeezzz tell us why ProMax is not working with FLUX? The Sampler gives me back: Error occurred when executing SamplerCustomAdvanced: mat1 and mat2 shapes cannot be multiplied (1x768 and 2816x1280)

Works in A1111. But Open-pose is very slow and need very high weights.

Laura what do you think about Flux? Fine tuning will work only with huge GPU ram.

Thank you . By the way what cloud GPU do you recommend to use ?

Thank you! ❤

Excellent video - cleared up any questions I had from your last one. Thank you, Laura!

Hi Laura, thanks a lot for the nice video :) Is stable-diffusion-v1-5 still the latest and best version to install?

I really loves, it's eye opening. I run an ecommerce website, do I will be using it to generate product mockup for my store. Can you please make a video on how to setup stable diffusion on deepnote. I just discovered that deepnote is free, but I'm not a programmer

Can you make a tutorial for A1111, I love your content 🫶🏻

thank u

Could have been longer. No coffee in the morning?

I LOVE YOUR VIDEOS AMAZING!!!

👋

Very cool! Thank you for uploading! Would love to see a video adding some sort of inpainting into this workflow. 🙏

not similar to original face

For me, training without regulation img make largely better results, also without captioning, I did a training with 100 regularization images in 34 passes and 68 reference images in 5 passes, and I ended up with a model that was completely off. I did a second test without regularization images and without captions, and in 13 minutes I got better results than a 2-hour training session. This is the third time that regularization images have just messed up my training. Maybe I misunderstood something, but it seems like they are just being used as training data, which is not the point, Anyway, thanks for the tutorial, it's the most comprehensive one on YT

Hey Laura, This is going to be a long one. And it is off topic. I just want to discuss your UA-cam performance. The quality of content, the personality (not many women doing this), and the viewer feedback (extremely positive) are way above what your view counts suggest. I think your thumbnails are a problem. 1. Try getting rid of that Square white frame. 2. Try having your hair loose/free and your face at least 1/2 the thumbnail. 3. For your skin and hair, I think a light background works better. (Keep background simple) 4. Keep the font / Title on the Left like you have it now. 5. Avoid the Cringe Expression Trend, but do try out Arrows.

FYI - This is available to insteall via ComfUI Manager's Model Manager now. This should save people an extra step or two.

I haven't still checked the documentation, does it work on A1111, and if so, which version of controlnet can run it? Thx!!

like and subscribe!

Thank you for your clear explanations. It's always a pleasure to watch your videos!

Thanks buddy, it works for me! Now I'll be having fun all day with this!

heyyyy new video! its nice to see you again, I gave up about that ai thing until I buy an Nvidia card but I will watch anyway :P

Thanks for sharing! I`m still new to Comfy. Now you can only use one model at the time with this particular workflow? I recall in A1111 we could stack them and use multiple controlnet how can this be done in ComfyUI?

can this work in FOOOCUS?

well, for arty effect on top of a dancer, very acceptable, now who will use this for other things that need consistency more than arty? no one.

This one is a game changer, all in one and takes so much less disk space

Thank you! Best video on using controlnet for SDXL!

Well, that's a lot better than downloading all those models

Openpos is not working well with that

You’re the Best! Thanks!

Thank you . Does it work on forge ui ?

Thanks for your presentation!