跑车/越野车/皮卡/赛车XL LoRA

使用介绍

这个LoRA的基本思想是制造宽体汽车。

当与汽车 LoRA 或可以吸引汽车的检查站一起使用时，它应该能够在汽车上添加某种宽体套件。

所有汽车分为 6 种不同类型：

• Sports Car
• Super Car
• Sedan
• Hatchback
• SUV
• Truck

建议权重：0.6~1.0

提示格式：颜色（可选） +宽体+体型（见上文）

开始的示例提示：

((masterpiece,best quality)), ultra-detailed, blue widebody sports car, vehicle focus,

关于此版本

• 基于 SDXL 1.0 基础模型和 FP16 进行训练
• 扩展所有类别的数据集
• 重新标注所有数据，包括两个摄像机角度：from side, from behind
• 与其他汽车 Loras (SDXL) 结合使用时应像 SD1.5 一样工作

触发词：	widebody sports car widebody super car widebody hatchback widebody sedan widebody suv widebody truck
使用提示：	Strength: 0.8

作品参照

((masterpiece,best quality)), ultra detailed, widebody sports car, vehicle focus, from above,  <lora:WBC-XL-v1-000006:0.9>
Negative prompt: (worst quality, low quality), lowres, (bad), error, jpeg artifacts, watermark, signature, extra digits, username,
Steps: 40, CFG scale: 7, Sampler: DPM++ SDE, Seed: 1925831427, VAE: sdxl_vae.safetensors, Size: 832x1216, Model: [SDXL]Juggernaut_X_RunDiffusion, Version: v1.9.3, Model hash: d91d35736d, Hires upscale: 2, Schedule type: Karras, Hires upscaler: 4x-UltraSharp, Denoising strength: 0.4, Downcast alphas_cumprod: True, Clip skip: 2

((masterpiece,best quality)), ultra detailed, light grey widebody hatchback, in a empty outdoor parking lot, sunlight, depth of field, reflection,  <lora:WBC-XL-v1-000006:0.9>
Negative prompt: (worst quality, low quality), lowres, (bad), error, jpeg artifacts, watermark, signature, extra digits, username,
Steps: 40, CFG scale: 7, Sampler: DPM++ SDE, Seed: 1472749203, VAE: sdxl_vae.safetensors, Size: 1216x832, Model: [SDXL]Juggernaut_X_RunDiffusion, Version: v1.9.3, Model hash: d91d35736d, Hires upscale: 2, Schedule type: Karras, Hires upscaler: 4x-UltraSharp, Denoising strength: 0.4, Downcast alphas_cumprod: True, Clip skip: 2

((masterpiece,best quality)), ultra-detailed, 4k, highres, cinematic, widebody sedan, vehicle focus, city, neon lights,
Negative prompt: bad-artist, (low quality, worst quality:1.4), watermark, username, artist name, signature,
Steps: 30, CFG scale: 8, Sampler: DPM++ SDE Karras, Seed: 2710898877, Size: 512x704, Model: realisticVisionV20_v20NoVAE, Version: v1.3.2, Model hash: e6415c4892, Hires upscale: 2, AddNet Enabled: True, AddNet Model 1: conceptWB_v1-000003(b08495ddf859), Hires upscaler: SwinIR_4x, AddNet Module 1: LoRA, AddNet Weight A 1: 0.8, AddNet Weight B 1: 0.8, Denoising strength: 0.5, Clip skip: 2