Technically Color WAN 2.2 is meticulously crafted to capture the unmistakable essence of classic film.
This LoRA was trained on approximately 100+ stills to excel at generating images imbued with the signature vibrant palettes, rich saturation, and dramatic lighting that defined an era of legendary classic film. This LoRA greatly enhances the depth and brilliance of hues, creating more realistic yet dreamlike textures, lush greens, brilliant blues, and sometimes even the distinctive glow seen in classic productions, making your outputs look truly like they've stepped right off a silver screen.
v1 - Trained for 3750 steps, this version is exclusively for use on still images and will likely not perform well for T2V because video clips were not trained on, use it for text to image tasks only. I'm uploading both the high and low noise variants.
A workflow is attached to the gallery images. The image results were generated with an RTX Pro 6000 and take around 4 minutes for each result. Lowering the resolution and eliminating sampler phases might improve the time, didn't try as quality was my main motivation.
Description
FAQ
Comments (8)
Huge thank you for this! I've been doing this look in post with Resolve but having the Lora to get it right out of the gate is super helpful.
Steve , how do I run this model, can you give me help 🙏 on how to
It looks like you misnamed your high-noise model low-noise, and the low-noise model is called high-noise. Once I figured this out, it works perfectly. Awesome LoRa!
Hmm, I don't think they're misnamed because AI Toolkit automatically named them. I wonder what happened there.
@renderartist That is strange. In my workflow, they work the opposite way to other high/low LoRas. I use phr00t's AIO v10, so I usually only use low noise LoRas, as the high noise ones result in a fuzzy mess. In this case the 'low noise' LoRa makes a fuzzy mess, and the 'high noise' produces the desired result.
+1 to this, using only the low model in my workflow.
dude is this t2v or i2v?
Amazing workflow. How do you train Wan loras and where?