[LoRA] D-mecha
模型简介：使用了59张DALL-E3生成的机甲图片进行训练，推荐权重0.9
59 mecha pictures generated by DALL-E3 were used for training, A weight of 0.9 is recommended.
If you like this model, you are welcome to support me here.
~~~~~~~~~~~~~~~~~~~~~~~o(*￣︶￣*)o~~~~~~~~~~~~~~~~~~~~~~~
主图提示词和参数：
Prompt: ((masterpiece)), (((best quality))),
<lora:daem4:0.9>,
(daem style:1.3), (mecha:1.3),(ink painting:1.3),monochrome, greyscale, from below,
solo, armor, horns, glowing, no humans, upper body, black and gold body, mechanical horns, black and gold mecha, tokusatsu, science fiction, grey background, red eyes, glowing eyes, blurry, simple background,,upper body, looking at viewer, facing viewer,
Negative prompt: EasyNegative, extra fingers,fewer fingers,(low quality, worst quality:1.9), (bad anatomy), (inaccurate limb:1.2),bad composition, inaccurate eyes, extra digit,fewer digits,(extra arms:1.2),(spider:1.3),,(holding:1.5),,(weapon:1.5),(gun:1.5),(gatling gun:1.5),
Steps: 20, VAE: orangemix.vae.pt, Size: 696x960, Seed: 2276898655, Model: DreamboxMix-A-V2.01protype, "daem4: 30fc76e80156", Version: v1.6.0, Sampler: DPM++ SDE Karras, VAE hash: f921fb3f29, CFG scale: 7, Clip skip: 2, Model hash: 89bec2f579, Hires upscale: 1.7, Hires upscaler: R-ESRGAN 4x+ Anime6B, Denoising strength: 0.52
页:
[1]