Draft_Kami_V1

Professional Cinematic Lighting & Texture Enhancement Model

Description:

This model is a specialized fine-tuning designed to enhance skin textures, micro-details, and complex lighting scenarios in photographic generations. It works best when applied to portraits and architectural visualizations where shadows and highlights need a more natural transition.

Our goal with this specific version is to reduce the "plastic" look often found in standard AI outputs, providing a more gritty, realistic, and tactile feel to every pixel. It has been trained on a curated dataset of high-dynamic-range (HDR) photographs and studio-lit environments to ensure that the global illumination remains consistent across different prompts.

Recommended Settings for Best Results:

• Sampler: DPM++ 2M Karras or Euler a

• Steps: 25 - 35

• CFG Scale: 5 - 7.5

• Clip Skip: 2

• Negative Prompt: (deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation.

Feel free to post your results in the gallery below! I am looking for community feedback to improve the next iteration of this weight set. Every image posted helps me understand how the model handles different prompts and seeds.