Colorful Realistic XL 彩璃写实XL版本v1.0 (ID: 629298)

Colorful Realistic XL, originally driven by personal interest, has been fine-tuned using a total of 12,379 high-quality images, tagged with GPT4V, WD14, and COGVLM. This version is shared for free use by individual users and non-profit organizations only. The model allows various forms of merging (e.g., 0.5 + 0.5 or MBW) but is restricted to private use. Sharing or publishing the merged models in any form is prohibited.

Welcome to follow my Github (https://github.com/luoqxi). I will upload high-frequency tags to Github soon.

I mainly use ComfyUI, but I also provide WebUI parameters for reference. If you use WebUI, don't forget to enable ADetailer for face and hand correction.

Negative Prompt: worst quality, low quality, illustration, 3d, 2d, painting, cartoons, sketch, watermark,

Steps: 30,

Sampler: DPM++ 2M, Schedule type: Karras,

CFG scale: 6.5,

Size: 832x1216,

Model: ColorfulRealisticXL_CE_V1.Fp16,

Clip skip: 2,

Issues: Due to material or training problems, this model tends to cause facial blurriness when generating full-body images. Therefore, face correction is crucial when generating images. You can solve this issue by yourself, and I will demonstrate this situation in the example images.

Please comply with relevant laws and regulations. When using this model, do not infringe on the legitimate rights and interests of others.

Colorful Realistic XL 彩璃写实XL，本是个人爱好驱动下制作，此版本微调总共使用了12379张高质量素材，打标为GPT4V，WD14，COGVLM三种都有,现分享仅供个人用户和非营利性公益组织免费使用，此模型不限制各种方式的融合（例如0.5 + 0.5 或者 MBW），但仅限私下使用，禁止将融合后的模型以任何形式分享或者发布,

欢迎Follow本人Github（https://github.com/luoqxi），后续我会把高频tag上传到Github,

本人基本只使用ComfyUI，但也给出相应Webui出图参数参考，你如果使用Webui不要忘记开ADetailer修脸和修手,

Negative Prompt: worst quality,low quality,illustration,3d,2d,painting,cartoons,sketch,watermark,

Steps: 30,

Sampler: DPM++ 2M, Schedule type: Karras,

CFG scale: 6.5,

Size: 832x1216,

Model: ColorfulRealisticXL_CE_V1.Fp16,

Clip skip: 2,

存在的问题：本模型因为素材或者训练的问题，在出全身图的时候容易导致脸部模糊，所以在出图的时候修脸就很重要,你可以自行解决这个问题，我会在例图中展示这个情况,

遵守相关法律规定，在使用本模型时不得侵害他人合法权益,

描述:

训练词语: