Version: Illustration:

• LoRA Strength: 1.20

• Model: flux1-dev-fp8

• Steps: 20

• CFG Scale: 3.5

• Sampler: DPM++ 2M SDE Heun Karras

• KSampler: dpmpp_2m_sde_gpu

• Schedule: karras

• Guidance: 3.5

• VAE: Automatic

• Clip Skip: 2

V1:This Lora doesn’t need any trigger words or keywords like illustration, digital art, or anime. It creates subjects ranging from semi-realistic to hyper-realistic.

V2: This LoRA was trained on Civitai using a dataset of 21 images, over 7 epochs, with around 25 repeats.