WAN 2.2 LORA Image Training Guide: Local & Runpod
Learn how to train WAN 2.2 LoRA for images both locally and on RunPod! In this step-by-step guide, I’ll show you how to set up the tuner, prepare your datasets, and train your own character LoRAs. Whether you’re using a powerful local GPU or renting one in the cloud, this tutorial will give you everything you need to get started.
💡 What you’ll learn:
▸ How to train LoRAs on WAN 2.2 for images
▸ Setting up locally with SK Tuner
▸ Training LoRAs on RunPod using my ready-to-use template
▸ How to install models (VAE, text encoders, diffusion models)
▸ Training both high-noise and low-noise models for best results
▸ Tips on parameters, learning rate, and epochs
📥 Resources & Downloads:
▸ RunPod Template: https://console.runpod.io/deploy?temp...
▸ RunPod: https://runpod.io?ref=49tc28ho
▸ Musubi Tuner GitHub: https://github.com/kohya-ss/musubi-tuner
▸ Workflows & Files (Free on Patreon):https://www.patreon.com/posts/wan-2-2...
✨ Support the Channel:
▸ Patreon: / endangeredai – Access exclusive content & early releases
▸ BuyMeACoffee: https://buymeacoffee.com/endangered – Support ongoing tutorials
▸ Newsletter: https://endangeredai.com/sign-up/ – Stay updated with the latest AI tools & workflows
🔈 Join my Discord: / discord
🌍 Follow me here:
Instagram: / endangeredai
TikTok: / endangered.ai
X (Twitter): https://x.com/endangered_ai
Facebook: / endangeredai
Website & Newsletter: https://endangeredai.com
⏱️ Chapters
0:00 Intro
0:22 Why WAN 2.2 LoRA for images
0:50 Local training setup
1:24 RunPod training setup & template
2:18 Installing SK Tuner
3:00 Downloading models (VAE, text encoders, diffusion)
3:30 Training workflow (high noise & low noise)
7:48 Training in practice
17:33 Support, Patreon, and next steps