Hello everyone!

As with my previous model (Pink_Auralis), this merge combines two great checkpoints: Auralis by @FlareAI_Studios tudios and RAMTHRUST'S-NSFW-PINK-ALCHEMY-MIX✨ by @RAMTHRUST.
This time I tweaked the recipe a bit, aiming for a more wild and colorful style while (trying to) preserve sharpness. Plus, I added a little extra spice, hehehe...

As always, I'm still a noob, and this merge is the result of blind experimentation and pure luck. However, I fell in love with some of the outputs, so I decided to share it.

It's capable of decent generations, but you'll need to find the right approach — the model won't do all the work on its own, unfortunately.

MY GENERATION SETUP

The model performed well in my basic workflow. I didn't encounter any serious quirks like with my first merge (where we had to use CFG=2 half the time — that was weird!).

First Pass (Base Generation):

• CFG Scale: 5.

  • I haven't even tried other values; results were quite good out of the box. There might be a better sweet spot though — I'd be happy if you experimented!

• Sampler/Scheduler: I use res_3m and beta57, but standard euler with normal or beta should also work fine if you prefer to keep it simple.

Upscaling (I use Latent Iterative):

• CFG Scale: 4.

  • Again, I haven't tested other values — it performed as expected.

• Sampler/Scheduler: dpmpp_2m with beta worked well for me. I haven't tested other combinations.

PROMPTING BEHAVIOR

Please note that the model is highly responsive to your prompt. If you use something like ultraviolet light, expect the image to be very ultraviolet. At the same time, it can be quite random — without a highly descriptive prompt, results will vary significantly from seed to seed. I find this feature kind of cool, but that's a matter of taste.

I also want to point out: this isn't the sharpest merge on the platform. It won't give you flawless five-fingered hands or razor-sharp silhouettes — honestly, I get distorted fingers quite often. If you're looking for clinical precision, you might be in the wrong place.

BUT! It's quite colorful, contrasty, and artistic. That's exactly what I was going for. If that sounds like your thing, give it a try — this model can produce some really cool results!

MY USUAL PROMPTS

• Negative Prompt: (bad quality:1.1), (worst quality:1.1), bad details, (bad hands:1.2), child, loli, pubic hair, shiny skin

• Positive Prompt: I usually start with masterpiece, best quality, newest, detailed eyes, colorful eyes. Then I add either photorealistic or anime coloring, cel shading depending on the style I want — or sometimes neither. There are no strict rules here, and I'm sure you'll find much better ways to prompt it! Tags like colorful, high contrast often look cool, so give them a try!

Feel free to experiment! I'll be thrilled if you manage to create something you like with it.

Have a good day :3