Trained on Voyager Spirit art (insta & website)

Krea2

Showcase images :

• Turbo model

• 768x1152 - x1.5 Hires (10 steps, 0.35 with 4x_NMKD-Siax-200k)

• Euler / simple / 10 steps / CFG 1

• Lora strength 1.1

Ideogram 4

Klein 9B

Showcase images :

• Distilled model

• 768x1152

• Euler / normal / 4 steps / CFG 1

• Lora strength of 1 - 1.2

• When upscaled : x1.5 - Hi-res 4 steps - 0.35

Qwen-2512

It was trained on 2512, but I prefer to use the original Qwen-Image model for inference.

Showcase images :

• Original Qwen-Image (BF16)

• Lightning lora

• 5 steps / 1 CFG

• DDIM & Beta

• Lora strength 0.8-1

ZIT

Samples were made :

• 10 steps / 1 CFG

• DDIM & Beta

• 1056*1584

• Lora strength of 0.8

Qwen

Samples were made with either :

• Qwen-Image (BF16) or Qwen-Edit (the first one) - Results are very different with the model you use

• Lightning Lora @ 5-10 steps / 1 CFG

• DDIM & Beta