AUTOMATIC1111 FULL TUTORIAL - Text to Image with Stable Diffusion
Вставка
- Опубліковано 7 сер 2024
- Here's a tutorial on A1111, a 100% free & open source tool to generate images using Stable Diffusion. We'll focus on the txt2img for this video.
0:00 Intro
0:33 Updating to latest version
1:02 Checkpoints
Civitai: civitai.com
2:17 Tabs & Settings
2:50 Prompting
Positive: 1girl, brown hair, wearing black shirt, 8K, masterpiece, absurdres, intricate details, best quality, sunny, city
Negative: easynegative, paintings, sketches, worst quality, lowres, normal quality, blurry, monochrome, grayscale, extra fingers, fewer fingers, extra limbs, deformed limbs, bad anatomy, disfigured, nsfw
5:08 Seed
6:23 Sampling method
7:19 Sampling steps
8:30 Width & height
8:46 Batch count & size
10:00 CFG scale
10:48 Hires fix & refiner
14:07 LoRAs
Discover thousands of AI Tools. Also available in 中文, español, 日本語:
ai-search.io/
Here's our equipment, in case you're wondering:
GPU: RTX 4080 amzn.to/3OCOJ8e
Mic: Shure SM7B amzn.to/3DErjt1
Secondary mic: Maono PD400x amzn.to/3Klhwvu
Audio interface: Scarlett Solo amzn.to/3qELMeu
CPU: i9 11900K amzn.to/3KmYs0b
Mouse: Logi G502 amzn.to/44e7KCF
If you found this helpful, consider supporting me here. Hopefully I can turn this from a side-hustle into a full-time thing!
ko-fi.com/aisearch - Наука та технологія
Easily, one of the best tutorials out there! Great explanation
Thanks!
Wow This tutorial is really clear! I'm excited for the upcoming ones. You've gained a sub! :D
Thanks for the sub!
one of the most simple to understand tutorials i found
thanks
Finally the full tutorial! I've been waiting for this.
Hope you enjoyed it!
Simple tutorial with no annoying loud music in the bg, thanks!
Thanks!
Man I love your work !! You’ve helped me so much so far !!
thank you!
Yoooo I tbh thought that short segment in ur previous video was enough, but thxn for this video 💪😎 You = Legend
Thanks!
Brilliant, thank you
Muito obrigado por mais um vídeo incrível
As a noob I really needed this! Thanks!
youre welcome!
Leaps and bounds in one session, thanks bro
No problem
wow thank you very much 💯
You're welcome!
I was subscribe for this ahaha ❤❤❤
Do the img2img too pls ❤❤❤
Thanks. Yes, coming soon
easynegative is a negative embedding, not just part of the prompt.
thank you bro now I know how to update automatic1111
nice!
easy negative is a textual inversion embedding.
Great tutorial, one of the best. I'm just confused about why you discourage the refiner. I use it and the images are much better.
Thx a lot
You're welcome!
can you explain whats the difference between comfyui and automatic1111, and which one should one use ?
thanks
You're welcome!
Hey man, what are the best paid services right now for using stable diffusion on the cloud? And which ones are good for training LORAs?
Thank you so much. Amazing tutorial.
It would be great if you could make video about comfyui and animatediff for the animation.
Thank you
Thanks! I'm playing around with animediff but it might be too complicated for our viewers. have you tried these other tools: ua-cam.com/video/tDidofPTwCc/v-deo.html
@ai-tools-search thank you, but I know all of those, and i use some of them. Actually, I usually work on animations using standard software, and AI is just a complement to that. ComfyUI seems quite powerful.
This is really helpful , you could've added some background sound
it feels a bit empty with only voice. nonetheless amazing video !
Thanks for the tip!
Thank for this awesome tutorial 🤩
ALL this can be done on the Google collab version too? By uploading your own Checkpoint and Loras?
Yes you can! But note that Google is cracking down on people using colab for image generation, so it might disconnect at times
@@theAIsearch thank you. And bruh, why Google is doing this, don't understand. Maybe they want to add something similar to their Google bard or something similar
Image generation takes a lot of computing power, so i assume google doesn't want too many people 'abusing' their servers for this purpose. It's best if you get a cuda GPU and run this locally
Is there a way to convert your own images into checkpoints?
hello i'm a newbie.. in 1:31 why is it 'style' section but not 'character' section?
Hello AI search can you tel me why mine on Sampling method don't have like your? Thanks
What's a checkpoint?
do you have a link to download this tool?
Hi, I have a question for you that might be a good idea for your next content.
I'm new to using Automatic1111 and the file size is getting bigger, around 40GB. I'm using ControlNet and have downloaded the extension or model seg, depth, etc. for ControlNet.
When I use depth and see the preview of the depth image, will it download again in the CMD section? I thought that once I had downloaded parts of ControlNet like depth, seg, etc., it wouldn't download again, but it turns out it still downloads other parts in CMD. This also applies to extensions. Is that how it works? Or will it only download the remaining parts in CMD when using depth, etc. for the first time?
it depends on the code of the extension. usually, it checks for requirements, and if you already have them installed, it won't install them again.
Thanks for explaining, it will be a question size. The images are mixed 2-3-4 times for the same image. What is the reason for this? Is there a setting?
what do you mean by 2-3-4? can you explain further
Thanks for this bro, I appreciate it so much. Do u think nvidia 2060 graphic card is fine for this? And Google Collab its still cracking down? I would prefer doing it there tbh
yes 2060 is fine. colab is still disconnecting for anything SD related
How does the seed change for the batch size/count? Does the seed change for other images after the first?
yes, lets say your seed is 1234, and your batch size is 3. the seeds for the 3 images would be 1234, 1235, 1236
Without GPU can we run this on raspberry Pi 4 model B guess how much times it takes to use it ❤
Ik it's not related to this but I made an rcv model through Google colab Page it was running fine days ago but it's says there's an error in my link or wtv can someone help me with that?!
Collab is cracking down on people using it for voice cloning, so it's best if you can run rvc locally
Hey I can't use sdxl with 4gb vram
yes 8+ GB is recommended
Ey man can u do a RVC 2 Tutorial i guess?
Sure, ill add it to my list! Have you watched this yet ua-cam.com/video/9TroP5mR3CM/v-deo.html
@@theAIsearchyep Already watch it looking forward on the RVC 2 Tutorial
How do I enable Stable Diffusion web UI in my PC browser?
download it here github.com/AUTOMATIC1111/stable-diffusion-webui
@@theAIsearch Does it require GPU to generate images/videos or it can be used without GPU?
Hello can you make gif images with this? i want the generated image to be my wallpaper thanks!
i think so, using the extension AnimateDiff
How does this compare with invoke ai?
They are very similar, but A1111 has more advanced features and extensions you can add
maybe now i can get this thing to stop producing constant nightmare fuel 😂
lol nightmare fuel
@@theAIsearch my mistake was thinking i could simply enter a description into the prompt without worrying about anything else just to get started & that i would get something like dall-e 3 with bing's image creator produces, & boy-howdy was i wrong 😅 seriously tho, thank you, this vid was the jumpstart i needed 🤘😎
What if We figured out how to Turn entire catalog of Pinterest into Checkpoints.
is it offline
yes
Do you know about ai audio enhancer
Can you elaborate?
How to use Automatic 1111 with no GPU :')
I have macbook air m2. So I use cloud provider like rundiffusion. They charge 0.5$ per hour.
@@ustatvofficial i have a good pc rn , thanks for the offer anyway
@@itschepi then why do you ask how to use it with no gpu
@@ustatvofficial look at the date of my comment
Use stable diffusion
there should be a different category for weeaboos on civitai..
Is this free or paid man ? I really cannot understand
It's free!
@@theAIsearchhow to install this on locally please guide me please
음성변조 소리는 들리지도 않고 설명만 드럽게 하네