Magic of Art (FLUX) - V1
推奨プロンプト
Abstract Lighting Effects, Hyperrealism, Midjourney Style, Photorealism, Cinematic, High Fidelity, Realism, Light Play
推奨ネガティブプロンプト
Worst quality, low quality, bad hands, bad body, extra digit, less digits, bad anatomy, missing fingers, poorly drawn face, poorly drawn hands, umbrella, nude, nsfw, text, letters, too many legs, deformed, too many fingers, (((2 heads))), duplicate, abstract, mutilated, toy, figurines, decorated, extra limbs, mutilated; bad art, poorly drawn, out of focus, strange colors, 2 heads, long neck, noisy, elongated body, blurry, cropped image, ugly, out of frame, rough draft, deformed hands, twisted fingers; double image, ugly arms, multiple heads, poorly rendered arms, missing limbs, distorted, cut off, oversaturated, underside, bad anatomy; poorly drawn face, mutation, mutated, floating limbs, severed limbs, long body, disgusting, grainy, extra fingers, crude proportions, missing arms; mutated arms, cloned face, missing legs
推奨パラメータ
samplers
steps
cfg
ヒント
Experiment with CFG scale values between 2.0 to 3.5
Experiment with steps between 14 to 22
Experiment with weights from 0.3 to 1, ideally between 0.6 to 0.8
バージョンのハイライト
TEST, TEST, TEST: experiment with the values
The LoRA model was created from 170 images taken from the public domain on the Internet. LoRA was trained on the FLUX model
CFG Scale: 2.0 - 3.5 experiment with the values
Steps: 14 - 22 experiment with the values
Weight: 0.3 - 1.
Works well with weight: ( 0.5 - 0.8 ), experiment with weights up to 1.2
The LoRA model was created from 170 images taken from the public domain on the Internet. LoRA was trained on the FLUX model
CFG Scale: 2.0 - 3.5 experiment with the values
Steps: 14 - 22 experiment with the values
Weight: 0.3 - 1.
Works well with weight: ( 0.6 - 0.8 ), experiment with weights up to 1
モデル詳細
ディスカッション
コメントを残すには log in してください。
