Intro to LoRA Models: What, Where, and How with Stable Diffusion

[@LaCarnevali]

🔥Upscale low-quality AI images to 4K/8K/32K with this easier AI tool: bit.ly/3XhreWC

[@zor2239]

Laura expalining LoRA 😳😳

[@Turbo187KillerB]

This is the first video that actually helped me understand what a LoRA is and how to use them. I've kinda had a hunch but I didn't really fully grasp them until your explaination. Thank you! I'm stoked to try LoRA models!

[@Aetherreal]

I've been generating with SD1111 for about 6 months and had no grasp at all between Loras and checkpoints - thank you so much for these extremely clear instructions 🙏

[@GengoSenmon]

Laura talking about Lora! Great job. Did not know about the useful AnyLora checkpoint. Definitely make that video about training our own Lora's.

[@roybranford3787]

Laura, I just cannot pay you enough thanks for this video. I started with StableDiffusion from the beginning. There were many KZbin vids that I watched and became totally confused as SD spiralled out of control. Your video has brought me down to Earth again. All of your concise explanations have registered with me and I now understand what's going on. Keep up the good work!

[@DeePatience]

youre an excellent teacher. i have absolutely no technical background and I'm able to follow and understand your instructions.

[@richardglady3009]

Yours is the first video I have seen that actually explains what LoRA is…thank you. Thank you for explaining and showing what you are doing. New subscriber.

[@fungus98]

I learned more from this video today just about how to properly run the ui than I have from all the stuff I've been reading, and I've been reading a lot

[@YouTubeName-qp8gs]

Everyone else giving tips seem to be three steps ahead Thank you so much for thoroughly explaining the basics :)

[@communityspiritualc704]

as per comment below...searched through heaps of videos and this was the first one that made sense of everything! Other videos I think are for more experienced users. Many Thanks !!!!😃

[@Dassy36]

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

[@a.f.warrick0404]

New to all of this and didn’t understand all the concepts. You explained very well along with examples. Thank you!

[@ditapeskova4]

I love how you explain things. This was so useful.

[@LeeDesignStudioYT]

That was the most easiest tutorial and explanation ever! ♥ Thank you so much! i was so confused how to use Loras vs checkpoints. You helped so much!

[@tonraynavy]

Thank you Laura, I know few KZbinrs who explain in as much detail as you do. You have a follower from Mexico.

[@PegaPorco]

Insanely good tutorial. I've watched several videos and you're by far the one that taught this best

[@WigfieldNicholas]

Great explanation. Ive found a lot of videos don't explain SD very clearly but this one does.

[@Mixobit]

Finalmente un inglese che si capisce!!! 😊 Mi stai insegnando parecchie cose. Grazie

[@anikiler]

Thanks Laura, your video helped me to use a Lora for the first time and understand it!!

[@serenamanganini1188]

thank youuuu. best tutorial about image generation ive seen so far

[@xt-cj7jg]

this is surprisingly a good video explaining how to use stable diffusion even if its only about Loras

[@fpvx3922]

Thank you very much... the other videos just do not do things step by step as you did. Thanks for helping a newbie :)

[@simpernchong]

Laura explaining LoRA. Thanks so much.

[@gamer_gex]

THANK YOU !!! THE CLEAREST EXPLANATION YET !!!

[@cgstone30]

Please make a video simplifying how to train a custom lora

[@zephilde]

Hello! Great tutorial! Could you answer some questions on LoRAs ? - Can you use LoRA on any models? You had shown diff between SDv1.5 and AnyLoRA but do we have to check the SD version of training or something else? - Can LoRA be negative? Like text inversion you can put in negative prompt? - Does it work with any methods: txt2img, img2img, in/ut-painting, controlNet ? Thx

[@LaCarnevali]

- You should use LoRA trained on 1.5 with 1.5 checkpoints. - You cannot use LoRA in the Negative Prompt (source: github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Features#lora). Many are using LoRA in the positive prompt but with a negative weight (in theory, you cannot even use a negative weight) - Yes, it works with all of them.

[@zephilde]

@@LaCarnevali Thx ! :)

[@Marcus_Ramour]

many thanks for sharing - this is a brilliant explanation and got me up and running very quickly!!

[@digigoliath]

Laura introduces LoRA. Fantastic! TQVM!!

[@creativeleodaily]

While watching your video, I am rendering some shots in Stable Diffusion since past 48 hours... I was curious How I could use Ai power in my MotoVlogging... And just last night i came across LoRa models, I was curious about it's functions and damn KZbin red my mind 😅 and suggested your video 😊

[@kdzvocalcovers3516]

Excellente...concise and pleasant instructions...Thank You for all the hard work and time ,,not to mention..the dedication you invested into this great tutorial...from all of us to You..a special Thank You..Bravo!!!

[@peedubu]

Laura is an amazing explainer. So clear and succinct. Good Job! I learned so much. Subscriber for life.

[@myukichi]

Thank you very much for this intro video to LoRA. I'm new to this and just got stable diffusion setup just the other day. I was able to understand your instructions very well and was able to create the exact image I found from citivai. This is all very cool stuff and I can't wait to play with this.

[@robertgoldbornatyout]

Laura talking about Lora! amazing, thankyou the tutorial , it was very helpful , you have such a lovely kind personally. you remind me of a very close lady friend i once knew her name was also LAURA , sadly she was one of the first to get covid and died, i have a song that i play when i think about her, its called Laura sung by Matt Monro, from now on when I play the song I will also think about you as well , Laura when you get a minute please go and play the song .💐💐💐

[@LaCarnevali]

👀👀👀👀

[@xxxxxNeMxxxxx]

Nice Introduction to get an overview about LoRas. Thanks!

[@patrickarena2866]

Buongiorno Laura! Really appreciated your video. I just installed Easy Diffusion last Sunday and got lost until I found your videos. Now I can brag and say I understand a little about Generative AI. So thanks a bunch I got started. Now I need more so I brows((big)) cat. Would you please suggest the title of your video you discuss that? You surely already have! 😊😊 Thanks

[@Renzsu]

Thanks for your clear explanation! I just got into sd last week and I’m trying to get up to speed. This was very helpful!

[@dennissdigitaldump8619]

Thank you for explaining that the Lora's need the correct model. Been trying to do cartoonlike tutorials, but the Lora just does messy graffiti or noise as output.

[@LuppoDesign]

Thank you so much. Please, more videos about this. It would be nice to learn about to create model also.

[@arifkuyucu]

Hi dear Laure... I really thank you for clear and cool tutorial... what you wrote in to terminal about restart your stable diffusion?

[@misaonthefly]

Wow you actually explained everything I needed to know

[@belinhafernandes]

Your content is great. I am following you and subscribing everywhere. There are Ai that has models Lora installed and it's possible also to upload more models, but few explanations about what they are. Of course, I understood what they do but I wanted more than just picking and using models. As a newbie to AI, I need to know what I am doing! Thank you so much Laura.

[@WicCaesar]

Thank you so much for explaining things so clearly! Subscribed and will bingewatch!

[@Lykon]

Randomly bumped here. Awesome to see my models featured. Make sure you also use highres fix when you generate a subject you like.

[@LaCarnevali]

Ehy Lykon, love your models! Good to see you here ;)

[@MichaelSBearre]

This video was super helpful! I watched it two times, I look forward to more of your content!

[@dorkfactory]

I learned so much from this video. thank you so much!

[@Ronoaldo]

THANK YOU so much! This was an amazing video. Simple and effective explanation. Liked + subbed!

[@content1]

Hi one question please. Sometimes when you click the info it doesnt include the GENERATION INFO for the image. Is that normal or am I looking in the wrong place? Thank you.

[@panchi128]

The generation info is included at the poster/original author’s whim. So it may have just been that the person that posted the image didn’t fill the info out or didn’t want people to recreate it.

[@desidoodh71]

Hello Laura, thanks for another excellent video and that is easy to follow. I have a question. How do we use multiple Lora Models, So I want to create an image with multiple people using different Lora models and clothing styles. Can't seem to figure that out.

[@LaCarnevali]

Hi, only need to add trigger words for different lora, plus add the lora in the prompt ( )

[@kite7214]

Laura, you are fabulous. Just like Jason below, you were brilliant in telling me EXACTLY what I needed to do to use Loras in SD. Thank you so much. I am gonna Subscribe right now. 🙂

[@Jeremy54321]

This was a great video. Thanks for putting it together. Would love to see a video about how to build a LoRA model.

[@LaCarnevali]

kzbin.info/www/bejne/rpnSnYVsZ7Zlqrssi=8JIgfajUwyVqs8dr

[@davesuccess]

Best tutorial on LoRA's, thank you!

[@alienj1208]

This is so helpful! Thank you so much for sharing!

[@KROQER]

wow miss carnevali! i learned more from you in 22 mins then the last 6 months!

[@kevinehsani3358]

Excellent explanations, best one on how to use stable diffusion, all other videos leave out far too many essential small details

[@heijyokada]

Thank you Laura! You're amazing!

[@kindoblue]

Just subscribed, as your videos are packed with down to earth explanations. Bravissima

[@Acidburn007]

you are a great techer... i learned a lot from this video,,subscribed

[@UBik0013]

very nice, very nice the tutorial also.. :D anyways, thank you for making this video. I finally managed to wrap my head around on to how to use loras with my invokeAI

[@BrianLife]

" lik emajic" Love it! Thanks so much, you are spreading Vulable info; I really enjoy your way of expressing too! Bless!

[@stumbleio]

Excellent explanation!! Thank you!!

[@setheheart4911]

You did an amazing job explaining everything. Please make more content. I just subbed!

[@SpartacoBaldelli]

Brava Laura. Finalmente ho capito per bene come si usano i Lora Models. :)

[@viperking6573]

Grazie mille per il video Laura 👏🏻👏🏻👏🏻

[@jonathanvilario5402]

This was so helpful! Thank you! Any chance you can make a video on textual inversion next?

[@LaCarnevali]

yep, can do that

[@Animus141]

Excellent tutorial, very helpful - thank you.

[@digital_magic]

great, thanx for the valuable info 🙂

[@QaamansLand]

Hola Lora, Do I need to download the Any-Lora Checkpoint if I already use stable diffusion? Thanks for your time. ~~Qaaman~~

[@LaCarnevali]

If you want to use Any-Lora checkpoint, yes, you need to download it

[@tiggersaldy6237]

Thanks for the clear video

[@greenrabbit4075]

at this point I am always thinking: "Okay..is this real? oO" Same with Laura ^^

[@NomadicJurassic]

thanks Laura. great work! I think theres a notebook that has a batch folder for LORA's. if you you find it before me let me know!

[@mjanich8]

Molte Gracie. Yes, would love to see how to train my own Lora model.

[@Bigblackninja5]

Is the anylora still better to use? And in the video I didnt see the anylora file in your folders where did you put it? Also on the anylora theres a lot more versions now im not sure which one do use (I have 8gb card)

[@gabbakopf9472]

Thank you so much, so straight to the point and easy to follow! Still haven't managed to make it work tho because the interface of Kohya for me looks different than in your video, so I end up with errors both on Windows and Mac. Would really appreciate an update on this!! :)

[@LaCarnevali]

Can try with this: kzbin.info/www/bejne/qovNnYemib9rgsU

[@tulipenoire369]

Thank's for explaining !

[@bodhimulasatyajati4410]

My Lora model is not loaded in the LORA tab although I have put LORA files in LORA Folder, what happen and what should i do?

[@UnarmedPug]

This was very helpful, thank you

[@isaacnaughton5206]

Great video on LoRas by Laura.

[@TucTuz]

honestly, you are amazing

[@SK-gi3gf]

Brilliant video - Thanks!

[@fratoguz4133]

Please shoot more videos, I can keep listening you forever

[@LaCarnevali]

😂

[@Goofer]

Thanks, this was very useful.

[@valentinacasale1715]

Very nice video, everything is very clear and well explained! If you don't know what to do next video about, I suggest UPSCALING 😅🤣

[@danielevolpe1086]

Grazie per i tuoi video! Molto ma molto utili!

[@SouthbayCreations]

I was just wondering where you were 😅. Great video! Thank you 🙏

[@Dannysingh-r7b]

Using chair and sofa images, I have finetuned a stable diffusion model. However, the model does not generate a living room using the chair and sofa. Please help me !!!!

[@rsteel1138]

Subscribed to your channel -- i've watched 3 vids of yours now; very helpful 💫

[@el.rascador]

So much information! Thanxs

[@ianm9853]

Thank you Laura!

[@simbarules777]

Thank you Laura, much appreciated, buen dia ;-)

[@IABGroup-fz6bv]

Very Clear. Thanks

[@petermolvaier]

great video; thank you. did you experiment with comfyUI on a mac? as i understand it is using better the resources.

[@LaCarnevali]

Yes, ComfyUI is better than SD on Mac, but still not as great as windows

[@faridmohammadizadch]

Good explanation, to the point

[@tecnopoesia]

Very instructive vídeo! Greetings from Brazil!

[@merce414]

Great video, thank you

[@MacShrike]

Very good explanation, thanks for the weight tip for multiple models, that had eluded me =)

[@aladinmovies]

Good explanation. Thanks

[@AlexandrGenin-q3k]

very nicely explained! Thank you!

[@pastuh]

Would be nice to get info about Dreambooth. So much info, it's hard to understand which best way to train model..

[@RSV9]

Is there a way to also save the Trigger Words so that I don't have to search every time in the place where the lora was downloaded ? I don't get the purple symbol under "Generate" even though I tried everything. I am using it in painting and it comes out quite well. Thanks Laura

[@RSV9]

But i have the Lora-Tab