-- RESOURCES USED --

Pinned below.

This is an AI model for pixel gears.

For the gears in the example images these following prompts were used.

Positive Prompts -- Gold Gear, carbon center, icon painting with a white background, casual game style, shading, minimalistic, <lora:GearV1:0.7>, <lora:PlixelSD15:0.6>,

Negative Prompts -- text, watermark, blurry, deformed, depth of field, realistic, 3d render, text, lines, symbols

I utilized mscenemix_v11.safetensors Stable Diffusion checkpoint with the vaeFtEma560000Ema_original VAE.

DEFAULT Size settings are 512 x 512 so the gears generate decently.

It appears through testing that the gears generate better with the Loras at the end of the prompt. I used my Gear model at 0.7 as well as :PlixelSD15 (for polishing) at 0.6.

EXTRA

__________________________________________________________________________________________________

Gold Gear, carbon center, icon painting with a white background, casual game style, shading, minimalistic, <lora:GearV1:0.7>, <lora:PlixelSD15:0.6>,
Negative prompt: text, watermark, blurry, deformed, depth of field, realistic, 3d render, text, lines, symbols
Steps: 35, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 345564734, Size: 512x512, Model hash: 28fba9c54a, Model: mscenemix_v11, VAE hash: 0b204ad0ca, VAE: vaeFtEma560000Ema_original.pt, Lora hashes: "GearV1: 897d0809d365, PlixelSD15: 2f11fb6466fa", Version: v1.6.1