ANIMA LoRA
■ Training Model = anima-base-v1.0
■ Operation Confirmation = NovaAnimeAMV40
■ Recommended Generation Environment
Generating Model = NovaAnimeAMV40
Sampling Method = Euler α
Schedule Type = Normal
Number of Sampling Steps = 30
Shift = 5
CFG Scale = 3
LoRA Weight = 0.8-1.2
■ Positive Prompts
score_8,highly detailed, amazing quality, very aesthetic, saturated,anime screencap,
■ Negative Prompts
score_1, score_2, score_3, blurry, jpeg artifacts3D,worst quality,normal quality,low quality,lowres,watermark,flat colour,monochrome,greyscale,bad hand,extra digits,fewer digits,censored,
*************************************************
■ Learning Model = illustriousXL_v01
■ Recommended generation environment
Generate Model = JANKU Trained+(v5.0 or v6.0) / WAI-illustrious-SDXL
Sampling method = Euler a
Schedule type = Karras
Sampling steps = 30
CFG Scale = 5
LoRA weight = 0.9
■ Positive prompt
highres, highly detailed, amazing quality, very aesthetic, anime screenshot,
■ Negative prompt
123qweasdz, 3D, deformation, (worst quality, low quality,normal quality), lowres, monochrome, grayscale, watermark, flat colour, no shading,
■ caption design
・Adding "anime screenshot" to the positive prompt makes the output closer to anime original art. Removing it slightly reduces character fidelity, but improves the quality of highlights, shading, and other texture details.
・"123qweasdz" in the negative prompt is strongly recommended. Due to the caption design of this LoRA, all non-character elements (background, clothing, props, etc.) are linked to this dummy token, so placing it in the negative prompt significantly improves generation stability.
■ Recommended resolution
21:9 = 1806 x 774
2:1 = 1672 x 836
16:9 = 1576 x 887
3:2 = 1448 x 965
4:3 = 1365 x 1024
1:1 = 1182 x 1182
