Home Creators Posts Import Register Favorites Logout

Content

Z-Image Turbo step by step: generate with ready SwarmUI presets, add Fun ControlNet Union and train LoRAs with Ostris AI Toolkit on GPUs from 6 GB.

1-click to download and install the very best Z-Image Turbo presets and models, including the 5.7 GB FP8 Scaled version for low VRAM. Includes full step by step LoRA training with configs for 8, 12 and 24 GB GPUs on Windows, RunPod and Massed Compute, as low as 6 GB GPU.

Full tutorial: https://www.youtube.com/watch?v=ezD6QO14kRc

Z-Image Turbo LoRA training with AI Toolkit and Z-Image ControlNet Full Tutorial for Highest Quality

Z-Image Turbo LoRA training with Ostris AI Toolkit + Z-Image Turbo Fun Controlnet Union + 1-click to download and install the very best Z-Image Turbo presets. In this tutorial, I will explain how to setup Z-Image Turbo model properly in your local PC with SwarmUI and download models and use them with highest quality via ready presets. Moreover, I will show to install Z-Image Turbo Fun Controlnet Union to generate amazing quality images with ControlNet preprocessors. Furthermore, I will show how to 1-click install AI Toolkit from Ostris and train Z-Image Turbo model LoRAs with highest quality configs made for every GPU like 8 GB GPUs, 12 GB GPUs, 24 GB GPUs and so on. I did a massive research to prepare these Z-Image Turbo model training configurations.

👇 Links & Resources Mentioned:

Download SwarmUI & Models: [ https://www.patreon.com/posts/Download-SwarmUI-Models-114517862 ]

Ostris AI Toolkit (SECourses Version): [ https://www.patreon.com/posts/Ostris-AI-Toolkit-140089077 ]

Ultimate Batch Image Processing App: [ https://www.patreon.com/posts/Ultimate-Batch-Image-Processing-App-120352012 ]

SwarmUI with ComfyUI Backend Windows Tutorial: [ https://youtu.be/c3gEoAyL2IE ]

SwarmUI with ComfyUI Backend RunPod and Massed Compute Cloud Tutorial: [ https://youtu.be/bBxgtVD3ek4 ]

⏱️ Video Chapters:

00:00:00 Introduction to Z-Image Turbo Model

00:00:54 FP8 Scaled Version 5.7GB for Low VRAM

00:01:10 ControlNet Union with Z-Image Turbo

00:01:30 LoRA Training with Ostris AI Toolkit

00:02:00 Default vs Custom Training Preset Quality Comparison

00:03:00 RunPod Cloud Training Preview

00:03:40 MassedCompute Cloud Training Preview

00:04:16 Downloading Z-Image Models via SwarmUI

00:05:00 Z-Image Turbo Core Bundle & ControlNet Files

00:05:58 FP8 Scaled Model & Musubi Tuner Converter

00:07:13 Updating ComfyUI for Sage & Flash Attention

00:08:13 Updating SwarmUI & ControlNet Preprocessors

00:08:52 Updating & Importing Latest SwarmUI Presets

00:09:20 Generating with Quality 2 Fast Preset

00:10:48 Generating with Quality 1 Upscale Preset

00:11:35 Quality 1 vs Quality 2 Visual Comparison

00:12:13 Setting up ControlNet Input & Aspect Ratio

00:13:41 ControlNet Strength Settings & Canny Test

00:15:26 Using Depth Preprocessor with Z-Image

00:15:58 Coloring Lineart Drawings with ControlNet

00:16:58 Lineart Preprocessing Comparison

00:17:50 Ostris AI Toolkit Installation Prerequisites

00:19:12 Installing Ostris AI Toolkit on Windows

00:20:02 First Time UI Setup & Launching Interface

00:21:04 Loading Custom Training Configs

00:21:38 Creating a New Dataset Structure

00:22:24 Ultimate Batch Image Processing App Install

00:23:17 Dataset Prep Stage 1: Auto-Zooming with SAM2

00:26:08 Dataset Prep Stage 2: Resizing to Exact Resolution

00:28:12 How to Select Best Training Images

00:30:24 Importance of Emotions & Angles in Datasets

00:31:44 Z-Image Resolution & Aspect Ratio Rules

00:33:21 Configuring Training Parameters & Epochs

00:36:52 Resolution Impact on Training Speed

00:37:46 Starting the Training Job on Windows

00:38:39 Monitoring Training Progress & VRAM

00:39:43 Checkpoint Generation Settings

00:40:40 Resuming Training from Last Checkpoint

00:42:09 Training Speeds on RTX 5090 vs 4090 vs 3060

00:43:01 Training Quality: Default vs Custom Preset Comparison

00:44:21 Testing LoRAs with SwarmUI Grid Generator

00:46:04 Fixing ControlNet Error in Grid Generation

00:47:09 Comparing Generated LoRA Checkpoints

00:47:38 Using Trained LoRA with ControlNet Union

00:48:10 RunPod: GPU Selection & Template Setup

00:50:32 RunPod: Port 8675 Config & Initialization

00:51:36 RunPod: Uploading Installation Files

00:52:01 RunPod: One-Click Installation Command

00:54:07 RunPod: Starting AI Toolkit & Proxy Connection

00:54:38 RunPod: Uploading Dataset via Interface

00:55:32 RunPod: Starting the Training Job

00:56:24 RunPod: Speed & Cost Analysis

00:57:28 RunPod: Auto-Stop Command Setup

00:58:24 MassedCompute: GPU Selection & Coupon Code

01:00:16 MassedCompute: ThinLinc Client Setup

01:01:21 MassedCompute: Transferring Files to Shared Folder

01:02:55 MassedCompute: Installation Command

01:05:49 MassedCompute: Connecting via Public URL

01:06:54 MassedCompute: Starting Training Job

01:08:43 Downloading Checkpoints & Stopping Instance

🚀 Master Z-Image Turbo & LoRA Training: The Ultimate Guide!

In this comprehensive tutorial, I show you how to generate ultra-realistic images in seconds using the lightweight Z-Image Turbo model. We cover everything from 1-click installation on SwarmUI (ComfyUI backend) to mastering ControlNet Union for precise image control.

But that’s not all! I also reveal how to train your own high-quality Z-Image Turbo LoRAs using the Ostris AI Toolkit. I have developed a custom training preset that significantly outperforms the default settings—you have to see the comparison to believe it! Whether you are on a local PC, RunPod, or MassedCompute, this guide has you covered.

🔥 What You Will Learn:

Z-Image Turbo Setup: How to run this fast, 6GB model (FP8 included) on almost any GPU.

ControlNet Mastery: Use Canny, Depth, and Lineart to control your generations perfectly.

LoRA Training: Step-by-step guide

0

19


Related posts

All posts in this topic: LoRA and Fine-Tuning Training: Kohya, Musubi Tuner, OneTrainer, AI Toolkit · New to SECourses? Start Here: Best and Most Popular SECourses Posts

Files

Z-Image Turbo LoRA training with AI Toolkit and Z-Image ControlNet Full Tutorial for Highest Quality

Z-Image Turbo LoRA training with Ostris AI Toolkit + Z-Image Turbo Fun Controlnet Union + 1-click to download and install the very best Z-Image Turbo presets. In this tutorial, I will explain how to setup Z-Image Turbo model properly in your local PC with SwarmUI and download models and use them with highest quality via ready presets. Moreover, I will show to install Z-Image Turbo Fun Controlnet Union to generate amazing quality images with ControlNet preprocessors. Furthermore, I will show how to 1-click install AI Toolkit from Ostris and train Z-Image Turbo model LoRAs with highest quality configs made for every GPU like 8 GB GPUs, 12 GB GPUs, 24 GB GPUs and so on. I did a massive research to prepare these Z-Image Turbo model training configurations. 👇 Links & Resources Mentioned: Download SwarmUI & Models: [ https://www.patreon.com/posts/Download-SwarmUI-Models-114517862 ] Ostris AI Toolkit (SECourses Version): [ https://www.patreon.com/posts/Ostris-AI-Toolkit-140089077 ] Ultimate Batch Image Processing App: [ https://www.patreon.com/posts/Ultimate-Batch-Image-Processing-App-120352012 ] SwarmUI with ComfyUI Backend Windows Tutorial: [ https://youtu.be/c3gEoAyL2IE ] SwarmUI with ComfyUI Backend RunPod and Massed Compute Cloud Tutorial: [ https://youtu.be/bBxgtVD3ek4 ] ⏱️ Video Chapters: 0:00 Introduction to Z-Image Turbo Model 0:54 FP8 Scaled Version 5.7GB for Low VRAM 1:10 ControlNet Union with Z-Image Turbo 1:30 LoRA Training with Ostris AI Toolkit 2:00 Default vs Custom Training Preset Quality Comparison 3:00 RunPod Cloud Training Preview 3:40 MassedCompute Cloud Training Preview 4:16 Downloading Z-Image Models via SwarmUI 5:00 Z-Image Turbo Core Bundle & ControlNet Files 5:58 FP8 Scaled Model & Musubi Tuner Converter 7:13 Updating ComfyUI for Sage & Flash Attention 8:13 Updating SwarmUI & ControlNet Preprocessors 8:52 Updating & Importing Latest SwarmUI Presets 9:20 Generating with Quality 2 Fast Preset 10:48 Generating with Quality 1 Upscale Preset 11:35 Quality 1 vs Quality 2 Visual Comparison 12:13 Setting up ControlNet Input & Aspect Ratio 13:41 ControlNet Strength Settings & Canny Test 15:26 Using Depth Preprocessor with Z-Image 15:58 Coloring Lineart Drawings with ControlNet 16:58 Lineart Preprocessing Comparison 17:50 Ostris AI Toolkit Installation Prerequisites 19:12 Installing Ostris AI Toolkit on Windows 20:02 First Time UI Setup & Launching Interface 21:04 Loading Custom Training Configs 21:38 Creating a New Dataset Structure 22:24 Ultimate Batch Image Processing App Install 23:17 Dataset Prep Stage 1: Auto-Zooming with SAM2 26:08 Dataset Prep Stage 2: Resizing to Exact Resolution 28:12 How to Select Best Training Images 30:24 Importance of Emotions & Angles in Datasets 31:44 Z-Image Resolution & Aspect Ratio Rules 33:21 Configuring Training Parameters & Epochs 36:52 Resolution Impact on Training Speed 37:46 Starting the Training Job on Windows 38:39 Monitoring Training Progress & VRAM 39:43 Checkpoint Generation Settings 40:40 Resuming Training from Last Checkpoint 42:09 Training Speeds on RTX 5090 vs 4090 vs 3060 43:01 Training Quality: Default vs Custom Preset Comparison 44:21 Testing LoRAs with SwarmUI Grid Generator 46:04 Fixing ControlNet Error in Grid Generation 47:09 Comparing Generated LoRA Checkpoints 47:38 Using Trained LoRA with ControlNet Union 48:10 RunPod: GPU Selection & Template Setup 50:32 RunPod: Port 8675 Config & Initialization 51:36 RunPod: Uploading Installation Files 52:01 RunPod: One-Click Installation Command 54:07 RunPod: Starting AI Toolkit & Proxy Connection 54:38 RunPod: Uploading Dataset via Interface 55:32 RunPod: Starting the Training Job 56:24 RunPod: Speed & Cost Analysis 57:28 RunPod: Auto-Stop Command Setup 58:24 MassedCompute: GPU Selection & Coupon Code 1:00:16 MassedCompute: ThinLinc Client Setup 1:01:21 MassedCompute: Transferring Files to Shared Folder 1:02:55 MassedCompute: Installation Command 1:05:49 MassedCompute: Connecting via Public URL 1:06:54 MassedCompute: Starting Training Job 1:08:43 Downloading Checkpoints & Stopping Instance 🚀 Master Z-Image Turbo & LoRA Training: The Ultimate Guide! In this comprehensive tutorial, I show you how to generate ultra-realistic images in seconds using the lightweight Z-Image Turbo model. We cover everything from 1-click installation on SwarmUI (ComfyUI backend) to mastering ControlNet Union for precise image control. But that’s not all! I also reveal how to train your own high-quality Z-Image Turbo LoRAs using the Ostris AI Toolkit. I have developed a custom training preset that significantly outperforms the default settings—you have to see the comparison to believe it! Whether you are on a local PC, RunPod, or MassedCompute, this guide has you covered. 🔥 What You Will Learn: Z-Image Turbo Setup: How to run this fast, 6GB model (FP8 included) on almost any GPU. ControlNet Mastery: Use Canny, Depth, and Lineart to control your generations perfectly. LoRA Training: Step-by-step guide

Comments

Draxel

This time it feels like when Qwen came out, except the uptake on civitai is insane. We're in for SDXL round 2 I think.

Tyrell Burns

Seem to always have issues installing the sageatt wheel.

Legenderox Gaming

Hey Proffesor, I was wondering what do you think has best likeness for realistic photos. Flux SRPO, FLux2, Qwen, or ZImage? Lora and dreambooth ig too Thanks!

Leonid Wolf

Greetings. I wanted to know. Is there any way to speed up the lora training process on the 3060 ti video card? I set the training parameter to a resolution of 1024 x 1024, set 4000 steps on 20 images, and the training process took me more than 16 hours. Is there any way to get around this?

SECourses

yes. set resolution to 768px or 512px. it will be even faster

SECourses

100% depends on prompt. Qwen is the most strongest one for so many prompts. but if your prompts are suitable FLUX SRPO and ZImage will work great. I haven't trained FLUX 2 yet waiting Kohya to add Musubi repo to add your app hopefully

Pakmule

I could never get AI-Toolkit to run reliably, thanks for getting it working. Any chance you can come up with AI-Toolkit configs for other models? That was super easy to setup with your config settings.

SECourses

thanks a lot. i may do. currently i am researching wan 2.2 on our secourses musubi tuner premium app