Kakitsu/区域绘制工作流-Regional Prompter workflow—Densediffusion+Controlnet

English is not good！ 英语不好警告！ 注意：当使用多个角色+其他LoRA时，单次放大可能无法获得较好的效果。最好像我工作流一样 在放大完后再添加一个K采样生成一次。（可以是你原来的模型（应该？） 我是用的别的模型。如果生成或者放大出现报错 大概率是分辨率设置的问题 需要设置成64的倍数。 我在这里分享的所有图片都是原图，你可以拖入到你的comfyui查看tag以及其他信息。 tips：When using multiple LoRA models and generating multiple characters, a single upscaling pass may not yield optimal results. It's best to follow a workflow like mine: perform upscaling first, then connect a K-sampler for the final generation. If there is an error message, it should be due to the resolution setting issue. The resolution needs to be set to a multiple of 64. All the pictures I'm sh

<iframe src="https://silytics.si/embed/item/e7a333bf-dfa0-4687-828b-acae3cace8e4" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Civitai and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.