Home Creators Posts Import Register Favorites Logout

Content

Wan 2.2 complete LoRA training tutorial with Musubi Tuner, based on over 64 trainings; the ready configs train locally on GPUs with as little as 6 GB VRAM.

Covers text to image, text to video and image to video configs, dual high and low noise model training, hybrid image and video datasets, and generating with the trained LoRA in SwarmUI and ComfyUI. Training is shown on Windows and on the RunPod and Massed Compute cloud.

Full tutorial link > https://www.youtube.com/watch?v=ocEkhAsPOs4

Wan 2.2 Complete Training Tutorial - Text to Image, Text to Video, Image to Video, Windows & Cloud

Wan 2.2 training is now so easy. I have done over 64 different unique Wan 2.2 trainings to prepare the very best working training configurations for you. The configurations are fully working locally with as low as 6 GB GPUs. So you will be able to train your awesome Wan 2.2 image or video generation LoRAs on your Windows computer with easiness. Moreover, I have shown how to train on cloud platforms RunPod and Massed Compute so even if you have no GPU or you want faster training, you can train on cloud for very cheap prices fully privately.

📂 Resources & Links:

Download the One-Click Installer & Configs: [ https://www.patreon.com/posts/Musubi-Tuner-Trainer-App-Configs-137551634 ]

Qwen Image Model Training Tutorial (Prerequisite): [ https://youtu.be/DPX3eBTuO_Y ]

SwarmUI & ComfyUI Setup Guide for Windows: [ https://youtu.be/c3gEoAyL2IE ]

SwarmUI Installer and Model Downloader : [ https://www.patreon.com/posts/SwarmUI-Install-Download-Models-114517862 ]

ComfyUI Installer : [ https://www.patreon.com/posts/ComfyUI-Installers-105023709 ]

SwarmUI & ComfyUI Setup Guide for RunPod & Massed Compute: [ https://youtu.be/bBxgtVD3ek4 ]

Upload / Download Big Files Guide for RunPod & Massed Compute: [ https://youtu.be/X5WVZ0NMaTg ]

⏱️ Video Chapters:

00:00:00 Introduction to Wan 2.2 Training & Capabilities

00:00:56 Installing & Updating Musubi Tuner Locally

00:02:20 Explanation of Optimized Presets & Research Logic

00:04:00 Differences Between T2I, T2V, and I2V Configs

00:05:36 Extracting Files & Running Update Batch File

00:06:14 Downloading Wan 2.2 Training Models via Script

00:07:30 Loading Configs: Selecting GPU & VRAM Options

00:09:33 Using nvitop to Monitor RAM & VRAM Usage

00:10:28 Preparing Image Dataset & Trigger Words

00:11:17 Generating Dataset Config & Resolution Logic

00:12:55 Calculating Epochs & Checkpoint Save Frequency

00:13:40 Troubleshooting: Fixing Missing VAE Path Error

00:15:12 VRAM Cache Behavior & Training Speed Analysis

00:15:51 Trade-offs: Learning Rate vs Resolution vs Epochs

00:16:29 Installing SwarmUI & Updating ComfyUI Backend

00:18:13 Importing Latest Presets into SwarmUI

00:19:25 Downloading Inference Models via Script

00:20:33 Generating Images with Trained Low Noise LoRA

00:22:22 Upscaling Workflow for High-Fidelity Results

00:24:15 Increasing Base Resolution to 1280x1280

00:27:26 Text-to-Video Generation with Lightning LoRA

00:30:12 Image-to-Video Generation Workflow & Settings

00:31:35 Restarting Backend to Clear VRAM for Model Switching

00:33:45 Fixing RAM Crashes with Cache-None Argument

00:35:13 Dual Model (High & Low Noise) Training Setup

00:36:54 Preparing Hybrid Datasets (Images + Videos)

00:37:40 Manually Editing Dataset TOML for Resolution Control

00:39:53 Setting High Noise Model Paths for Dual Training

00:41:50 Optimization: Block Swap vs CPU Offload

00:43:10 Generating Video with Dual-Model Trained LoRA

00:45:35 Massed Compute: Server Setup & Coupon Code

00:47:00 Connecting via ThinLinc & File Transfer Methods

00:49:12 Massed Compute: Fast UV Installation & Downloads

00:50:27 Loading Configurations on Massed Compute

00:52:18 Troubleshooting: Fixing Config Version Error

00:53:20 Dual Model Training Speed Analysis on Cloud

00:55:40 RunPod: Selecting the Correct Template & GPU

00:57:45 RunPod: Uploading Files & Extracting Archive

00:58:38 RunPod: Terminal Installation & Model Downloads

01:00:26 RunPod: Correct Pathing Syntax & Backslash Fix

01:01:28 Setting Dataset Paths on RunPod

01:03:34 Installing nvitop on RunPod Terminal

01:03:54 Speed Hack: Disabling Numpy Memory Mapping

01:06:00 Terminating Instances & Final Remarks

100

Greetings everyone! Today I am presenting an epic tutorial on how to train the Wan 2.2 model to generate extremely high-quality, realistic images and videos. This is currently the most advanced model for generating life-like textures and details.

In this comprehensive guide, I cover everything you need to know to train Wan 2.2 on your local Windows computer, as well as on cloud platforms like RunPod and Massed Compute. We utilize the SECourses Musubi Tuner with fully optimized, 1-click presets designed for every GPU range (from 6GB to 192GB VRAM).

🚀 What You Will Learn in This Tutorial:

Wan 2.2 Text-to-Image Training: How to train the Low Noise model for massive detail and realism.

Wan 2.2 Text-to-Video Training: Mastering Dual Model training (Low Noise + High Noise) for superior video consistency.

Image-to-Video Workflow: How to use your trained LoRAs to animate static images.

Cloud Training: Step-by-step guides for Massed Compute (ultra-fast disk speeds) and RunPod.

Performance Optimization: Using FP8 scaling, Block Swapping, and CPU offloading to train on consumer GPUs.

Inference & Upscaling: Using SwarmUI and ComfyUI to generate and upscale content to 4K resolution.

💡 Key Features of Our Workflow:

Auto-Resume & Speed: New UV package installers for lightning-fast setup.

Presets for All GPUs: Configurations included for 6GB, 12GB, 24GB, 48GB, and 80GB+ cards.

Dataset Automation: Auto-resizing and captioning for both image and video datasets.

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

Wan 2.2 Complete Training Tutorial - Text to Image, Text to Video, Image to Video, Windows & Cloud

Wan 2.2 training is now so easy. I have done over 64 different unique Wan 2.2 trainings to prepare the very best working training configurations for you. The configurations are fully working locally with as low as 6 GB GPUs. So you will be able to train your awesome Wan 2.2 image or video generation LoRAs on your Windows computer with easiness. Moreover, I have shown how to train on cloud platforms RunPod and Massed Compute so even if you have no GPU or you want faster training, you can train on cloud for very cheap prices fully privately. 📂 Resources & Links: Download the One-Click Installer & Configs: [ https://www.patreon.com/posts/Musubi-Tuner-Trainer-App-Configs-137551634 ] Qwen Image Model Training Tutorial (Prerequisite): [ https://youtu.be/DPX3eBTuO_Y ] SwarmUI & ComfyUI Setup Guide for Windows: [ https://youtu.be/c3gEoAyL2IE ] SwarmUI Installer and Model Downloader : [ https://www.patreon.com/posts/SwarmUI-Install-Download-Models-114517862 ] ComfyUI Installer : [ https://www.patreon.com/posts/ComfyUI-Installers-105023709 ] SwarmUI & ComfyUI Setup Guide for RunPod & Massed Compute: [ https://youtu.be/bBxgtVD3ek4 ] Upload / Download Big Files Guide for RunPod & Massed Compute: [ https://youtu.be/X5WVZ0NMaTg ] ⏱️ Video Chapters: 0:00 Introduction to Wan 2.2 Training & Capabilities 0:56 Installing & Updating Musubi Tuner Locally 2:20 Explanation of Optimized Presets & Research Logic 4:00 Differences Between T2I, T2V, and I2V Configs 5:36 Extracting Files & Running Update Batch File 6:14 Downloading Wan 2.2 Training Models via Script 7:30 Loading Configs: Selecting GPU & VRAM Options 9:33 Using nvitop to Monitor RAM & VRAM Usage 10:28 Preparing Image Dataset & Trigger Words 11:17 Generating Dataset Config & Resolution Logic 12:55 Calculating Epochs & Checkpoint Save Frequency 13:40 Troubleshooting: Fixing Missing VAE Path Error 15:12 VRAM Cache Behavior & Training Speed Analysis 15:51 Trade-offs: Learning Rate vs Resolution vs Epochs 16:29 Installing SwarmUI & Updating ComfyUI Backend 18:13 Importing Latest Presets into SwarmUI 19:25 Downloading Inference Models via Script 20:33 Generating Images with Trained Low Noise LoRA 22:22 Upscaling Workflow for High-Fidelity Results 24:15 Increasing Base Resolution to 1280x1280 27:26 Text-to-Video Generation with Lightning LoRA 30:12 Image-to-Video Generation Workflow & Settings 31:35 Restarting Backend to Clear VRAM for Model Switching 33:45 Fixing RAM Crashes with Cache-None Argument 35:13 Dual Model (High & Low Noise) Training Setup 36:54 Preparing Hybrid Datasets (Images + Videos) 37:40 Manually Editing Dataset TOML for Resolution Control 39:53 Setting High Noise Model Paths for Dual Training 41:50 Optimization: Block Swap vs CPU Offload 43:10 Generating Video with Dual-Model Trained LoRA 45:35 Massed Compute: Server Setup & Coupon Code 47:00 Connecting via ThinLinc & File Transfer Methods 49:12 Massed Compute: Fast UV Installation & Downloads 50:27 Loading Configurations on Massed Compute 52:18 Troubleshooting: Fixing Config Version Error 53:20 Dual Model Training Speed Analysis on Cloud 55:40 RunPod: Selecting the Correct Template & GPU 57:45 RunPod: Uploading Files & Extracting Archive 58:38 RunPod: Terminal Installation & Model Downloads 1:00:26 RunPod: Correct Pathing Syntax & Backslash Fix 1:01:28 Setting Dataset Paths on RunPod 1:03:34 Installing nvitop on RunPod Terminal 1:03:54 Speed Hack: Disabling Numpy Memory Mapping 1:06:00 Terminating Instances & Final Remarks Greetings everyone! Today I am presenting an epic tutorial on how to train the Wan 2.2 model to generate extremely high-quality, realistic images and videos. This is currently the most advanced model for generating life-like textures and details. In this comprehensive guide, I cover everything you need to know to train Wan 2.2 on your local Windows computer, as well as on cloud platforms like RunPod and Massed Compute. We utilize the SECourses Musubi Tuner with fully optimized, 1-click presets designed for every GPU range (from 6GB to 192GB VRAM). 🚀 What You Will Learn in This Tutorial: Wan 2.2 Text-to-Image Training: How to train the Low Noise model for massive detail and realism. Wan 2.2 Text-to-Video Training: Mastering Dual Model training (Low Noise + High Noise) for superior video consistency. Image-to-Video Workflow: How to use your trained LoRAs to animate static images. Cloud Training: Step-by-step guides for Massed Compute (ultra-fast disk speeds) and RunPod. Performance Optimization: Using FP8 scaling, Block Swapping, and CPU offloading to train on consumer GPUs. Inference & Upscaling: Using SwarmUI and ComfyUI to generate and upscale content to 4K resolution. 💡 Key Features of Our Workflow: Auto-Resume & Speed: New UV package installers for lightning-fast setup. Presets for All GPUs: Configurations included for 6GB, 12GB, 24GB, 48GB, and 80GB+ cards. Dataset Automation: Auto-resizing and captioning for both image and video datasets.

Comments

abaj006

Do you currently have a Wan 2.2 text to image workflow for ComfyUI? I tried one available in the community, it uses the ClownShark sampler. I could not get text to image to work any other way. Thanks