1 Intro 2:14
2 Pre-requirements of this tutorial 1:04
3 1-Click installer for Automatic1111 Web UI 0:39
4 How to install Automatic1111 Web UI for SDXL and SD 1.5 models 0:14
5 How to checkout and verify your installed Python version 0:35
6 Which Automatic1111 Web UI command line arguments you need for SDXL 0:23
7 Where to find all command line arguments of Automatic1111 Web UI 0:29
8 Where to download SDXL model files and VAE file 0:31
9 Which folders you need to put model and VAE files 1:03
10 Detailed explanation of what is VAE Variational Autoencoder of Stable Diffusion 0:35
11 How to set your VAE and enable quick VAE selection options in Automatic1111 0:24
12 What does Automatic and None options mean in SD VAE selection 0:20
13 Why you shouldn't use embedded VAE of SD 1.0 model 0:54
14 Correct resolution of SDXL - how to use SDXL 0:19
15 How to install Kohya SS GUI script for SDXL training 2:30
16 What to do if your CMD is not progressing 0:49
17 When you need to use FP16 instead of BF16 0:35
18 How to install Kohya on RunPod or on a Unix system 0:39
19 How to start Kohya GUI after installation 0:42
20 What are Stable Diffusion LoRA and DreamBooth rare token, class token, and more training 0:26
21 How to select SDXL model for LoRA training in Kohya GUI 0:45
22 How to save and load your Kohya SS training configuration 0:09
23 How to use my own used configuration for this tutorial video training 1:14
24 How to prepare your training images for Kohya LoRA or DreamBooth SDXL training 1:20
25 What kind of training images you should use for training 1:39
26 What kind of regularization images you should use? The logic of using ground truth images 3:37
27 What is number of repeating in Kohya SS. Which number you need to pick 1:20
28 Where will be your LoRA checkpoints saved 0:34
29 How to verify your training images dataset properly composed 0:40
30 How to set your generated LoRA file names 0:33
31 Which training parameters you should use for SDXL LoRA training 0:09
32 Why select train batch size 1 and gradient accumulation steps 1 0:21
33 The logic of number of epochs 2:06
34 Detailed explanation of Kohya SS training. What each parameter and option do 0:37
35 Which learning rate for SDXL Kohya LoRA training 0:06
36 Why do I use Adafactor 1:28
37 The rest of training settings 1:16
38 Which Network Rank Dimension you need to select and why 0:47
39 How to fix if you get out of VRAM error - not enough memory 1:19
40 What is Network Alpha of Kohya LoRA 0:30
41 Don't forget Gradient Checkpointing 0:31
42 How to continue training with Kohya LoRA training 0:34
43 What does print training command do 0:16
44 How to calculate number of steps for each Epoch 0:12
45 How to calculate how many regularization images you need 0:30
46 When you should increase batch size when doing Stable Diffusion training? 0:43
47 How number of total steps max training steps are calculated in Kohya training 0:57
48 How you can generate your own regularization / classification images 1:19
49 How to manually edit generated Kohya training command and execute it 1:35
50 How to start training in Kohya 0:14
51 How to do training on your second GPU with Kohya SS 2:54
52 How much VRAM is SDXL LoRA training using with Network Rank Dimension 32 0:43
53 SDXL LoRA training speed of RTX 3060 0:09
54 How to fix image file is truncated error 0:39
55 How to reach and contact me if you get an error 0:44
56 VRAM usage and speed testing of different Network Rank 2:49
57 How to use absolute min VRAM to make it work 0:35
58 When is first checkpoint generated and where they are saved 0:55
59 How to continue training from saved state 0:42
60 Auto saved configuration files 1:46
61 How to use LoRAs with Automatic1111 Web UI 2:19
62 How to assign previews to your LoRA files / checkpoints 0:57
63 How to do x/y/z LoRA checkpoint comparison to find best LoRA model 4:09
64 How to understand if your LoRA model is overtrained / cooked or not 1:39
65 Testing our LoRAs stylization capability 2:16
66 How to generate studio shot quality images that you can use on LinkedIn, Twitter, Instagram and such 0:32
67 How to find best generated images with using an AI tool 4:01
68 How to utilize ChatGPT to find very good prompts 0:36
69 How to utilize high-res fix and LoRA inpainting to get amazing quality distant shot images 3:42
70 How to fix hands and face 3:49
71 How to use same training command I used 0:36
72 When you need to reduce weight / emphasis of the rare token 2:54
73 How to join our Discord community for help and tips