How Use Stable Diffusion, SDXL, ControlNet, LoRAs For FREE Without A GPU On Kaggle Like Google Colab

You want to use Stable Diffusion, use image generative #AI models for free, but you can't pay online services or you don't have a strong computer. Then this is the tutorial you were looking for. By watching this tutorial, you will learn how to use Kaggle free cloud service with famous Stable Diffusion #Automatic1111 SD Web UI as easy as it is running on your local computer. I have prepared an amazing Kaggle notebook that even supports SDXL and ControlNet of SDXL and LoRAs and custom models of #SDXL. Of course it supports all of the Stable Diffusion SD 1.5 based models too.

Kaggle Automatic1111 Notebook File ⤵️

Tutorial GitHub Readme File ⤵️

0:00 Introduction to how to use Stable Diffusion for free without a computer or a GPU
2:44 How to register a Free Kaggle Account and activate it
3:28 How to create a new Notebook on Kaggle
3:42 The two very important steps that you need to make before start using Kaggle Notebook
3:52 How to set accelerator (select GPUs) and enable Internet
4:28 What is persistence and should you use it
4:45 How to write code on a Kaggle notebook and use the code shared in the GitHub readme file
4:55 How to manually setup your Stable Diffusion Automatic1111 Web UI notebook if you are not my Patreon supporter
5:08 How to download and import the Automatic1111 SD Web UI notebook used in this tutorial
6:20 Suggested resources before following this tutorial
7:28 How to start your Free Kaggle Dual GPU session
7:39 How to see how many resources you are using in your current session in a Kaggle notebook
8:04 How to install Automatic1111 on a Kaggle notebook
8:43 Explanation of the right section of the Kaggle notebook, working directory
9:00 How to clear the outputs of a Kaggle notebook session
9:10 How to download Stable Diffusion SD 1.5, LoRAs and SDXL models into the correct Kaggle directory
9:39 How to download models manually if you are not my Patreon supporter
10:14 An example of how to download a LoRA model from CivitAI
11:11 An example of how to download a full model checkpoint from CivitAI
11:48 How to start downloading all the model files
13:41 How to install ControlNet extension and download ControlNet models
15:50 How to use custom LoRAs from CivitAI or Hugging Face
16:29 How to start Automatic1111 Web UI instance with correct parameters and settings
18:42 How to understand installation of Automatic1111 Web UI has been completed and ready to use
19:09 First time model loading may be very slow
19:26 How to enable quick VAE selection drop down list
19:50 How to set correct folder scan path for ControlNet
20:29 How to reload UI for extensions and setting changes to be effective
20:51 Automatic1111 Web UI is ready to use on a Free Kaggle notebook
21:22 How to use embedded VAE of the models
21:48 Which image generation sampler is the best one
21:58 Why and how much first image generation is slower
22:57 How to install extensions (e.g. After Detailer) of Automatic1111 on a Free Kaggle notebook
23:35 You need to reload Web UI to see new extensions
23:55 How to use your downloaded LoRA models in Automatic1111 Web UI
25:17 First image generation with SDXL model
26:16 First time image generation speed vs consequent images generation on SDXL
26:54 First image generation with the custom LoRA model from CivitAI
27:20 Image generation speed of SDXL when using LoRA
27:41 How to use your own trained LoRAs, models or LoRAs, models from your computer
29:03 How to import your files, datasets in to your current session on Kaggle
29:21 How to use files imported as data set such as LoRAs or models that you have uploaded as dataset
30:23 How to restart Automatic1111 Web UI on a Kaggle notebook
31:06 How to prompt your own trained LoRA
31:32 How to enable After Detailer (adetailer) extension to improve faces of Stable Diffusion generated images
33:01 Where are the SD generated images are saved on a Kaggle notebook
33:45 SDXL with LoRA image generation speed
34:20 How to use Stable Diffusion XL (SDXL) ControlNet models in Automatic1111 Web UI on a free Kaggle
35:05 Where to download SDXL ControlNet models if you are not my Patreon supporter
36:13 Notebook crashes due to insufficient RAM when first time using SDXL ControlNet and how I fix it
37:39 First image generation results of SDXL ControlNet
39:21 SDXL ControlNet development discussion topic
40:22 Possible bug with ControlNet
41:10 How to use Stable Diffusion 1.5 version based ControlNet instead of SDXL ControlNet
43:29 What happens if you exceed your assigned disk space on a free Kaggle notebook
44:35 First time using SD 1.5 ControlNet model on a SD 1.5 based Realistic Vision 5.1 model
44:55 How to fix runtime error when changing SDXL to SD 1.5 based model
46:42 How to use PNG info to get the prompt
47:57 How to download all generated images

[SECourses]

If I have been of assistance to you and you would like to show your support for my work, please consider becoming a patron on 🥰 ⤵

Technology & Science: News, Tips, Tutorials, Tricks, Best Applications, Guides, Reviews ⤵

Playlist of StableDiffusion Tutorials, Automatic1111 and Google Colab Guides, DreamBooth, Textual Inversion / Embedding, LoRA, AI Upscaling, Pix2Pix, Img2Img ⤵

[grzesiektg]

Brilliant, and I've spent so many hours to set it on my machine :D
Great job to us all! :)

[latent-broadcasting]

Amazing! Thanks, I'm learning a lot with your tutorials

[DJHUNTERELDEBASTADOR]

Gracias Furkan !!! se te agradece mucho!!! Saludos desde Bolivia

[theymightbedavis]

Thanks again for the great approach to use Kaggle. Could you please also do a tutorial on getting SadTalker to work in Kaggle? That way, we can make some of the characters we create in Stable Diffusion actually speak. SadTalker have a Google Colab notebook linked on their Git page, but if it could be in Kaggle, then it would be so much more efficient with Kaggle's 2 GPUs!

[ai_and_gaming]

Is possible to use deforum + controlnet too?
I tried to use another kaggle notebook with deforum, but it didn't recognize the controlnet even though it was installed

[shun_almya]

A very wonderful clip. But I have two questions. The first is, is there a way to upload Marge models, for example, to the huggingface site directly from Kaggle? The second question: If I download the Extensions and restart the session, will all the Extensions be deleted and I will have to download them again? Thanks

[bennguyen]

Coming from Midjourney, any thoughts on using Fooocus/SDXL? You don't have total control (ex. CFG values) but it does seem to help in learning how some controls affect the image?

Aside from the many cool tools (ControlNet Tile, Realistic Vision V5.1, After Detailer, Automatic1111, DreamBooth), I've also seen how things can be mixed together! For example, LyCORIS (LoCon and LoHa) can be easily mixed with LoRA!

BTW, any thoughts on Stable Diffusion vs Dall-E 3 / Bing Image Creator?

[harshitpruthi]

i am having a little problem . after running all codes when i run the last code that is supposed to give the link for the interface . my notebook automatically gets disabled and it shows "error . terminated by user " in the end . the codes were working 1st time but after that this problem is arising . what should i do

[ashd]

Hi Dr. Is there Dreambooth level training for SDXL? I want to make some portfolio images. Or shall I follow your Dreambooth tutorial instead?

[foreverkripto]

I don't have a strong GPU, my graphic driver is intel silicon, that's why I keep getting this "Torch is not able to use GPU; add --skip-torch-cuda-test to COMMANDLINE_ARGS variable to disable this check" error. Do you think can I use this method instead?

[biitqef]

SECourses Hi, is it somehow possible to upload your ckpt or safertoon templates into the beautifully crafted notebook you made?

[dauxgaming]

once i instal this do i have to run this process of downloading runing and downloading every time

[DTTR]

This is wonderful! It works well. Thank you for taking the time to create this detailed and helpful tutorial. However, there does not seem to be enough RAM for the SDXL base refiner to run no matter what I try to do. Can you confirm that using the refiner is not possible with Kaggle?

[Abed-izuy]

Merhaba Dr. Furkan. Is this notebook scalable to get SVD (Stable Video Diffusion) to work on it? Or does it need new coding in a new Kaggle notebook?

[thinker]

That was amazing tutorial.❤ Can you make a tutorial on how to run comfyui in kaggle

[goktugyenisen]

Hocam, bu güzel paylaşımınız ve çalışmanız için teşekkürler. Bir Türk'ten ingilizce ders dinlemekte oldukça garip oldu. :) Yeni denk geldim kanalınıza bundan sonra sizi takipteyim. İyi çalışmalar :)

[Charkel]

I stopped my notebook and when i relaunched it it re installed all the extensions but not models. Can i make it persistent?=

[alisyr]

It is not possible to download the notebook from the link. Subscription to the site is required for five dollars

[crazykim]

Turns out that if you even think about generating something even remotely NSFW, Kaggle will indiscrimately, and immediately lock your account. Would have been good to know.
I was just trying out if the generator worked and if I had made any mistakes in setting it up, and had copied a few of the loras from my google colab notebook, only of which one was of NSFW nature, generated four SFW pictures without any loras added, session ended and my account got locked. Only took 15 minutes from starting the session. *Sigh*