I’ve been experimenting with WAN 2.2 for a few days now, and I’m excited to share my discovery. By chaining together three samplers and applying lightx2v after the first pass, I’ve seen a massive improvement in results with very minimal additional time to render.
This method was inspired by a thread on Hugging Face, where users discussed the benefits of triple chained samplers. I decided to give it a try, and the results have been astounding.
If you want to try it out for yourself, I’ve shared my flows on CivitAI. My I2V and T2V v1.2 workflows use this method, and I’m excited to hear about your experiences.
Triple chained samplers are a game-changer for anyone working with WAN 2.2. With minimal additional time and effort, you can achieve significantly improved results. Give it a try and see the difference for yourself!