Lily-IL Genjitsu

High-Quality, Realistic Image Generation Model

Model Capabilities:

Lily-IL Genjitsu is built to deliver ultra-realistic, high-resolution visuals with refined aesthetic detail. It is well-suited for a wide range of creative outputs, especially character-focused compositions. The model supports both stylized and naturalistic results with advanced control over lighting, texture, and expression.

Recommended Parameters:

• Steps: 20–25

• Sampler: DPM++ 2M

• CFG Scale: 5–7

Prompt Compatibility:

• Supports Booru-style and natural language prompts

• Fully compatible with character and artist tags

• Seamless integration with additional LoRAs for fine-tuned results

Quality Prompt (Recommended):

(masterpiece, best quality, ultra-detailed, absurdres, very aesthetic, high res)

Negative Prompt (Recommended):

modern, recent, oldest, cartoon, graphic, text, painting, crayon, graphite, glitch, deformed, mutated, ugly, disfigured, long body, lowres, bad ???????, missing fingers, extra digit, fewer digits, cropped, very displeasing, (worst quality, bad quality:1.2), sketch, jpeg artifacts, signature, watermark, username, chibi, loli, simple background, conjoined, ai-generated

Contact:

For testing, collaboration, or support, please reach out via Discord: @LucyMocchi