f2k9b One-Click Workflow to Real Person Conversion (No LoRa Required) - v1.0 Showcase

2K One-Click Workflow to Real Person Conversion (No LoRa Required)

First Post by a Newbie

Many workflows for converting to real person rendering are quite complex, some even requiring LoRa. After testing, LoRa proved ineffective; the only truly effective node is the "Reference Latent" node. Some workflows lack control over the ratio, making it impossible to adjust the redraw magnitude.

I prefer a minimalist approach, using almost entirely the nodes provided by ComfyUI. You can replace gguf with safetensors, and replace the image saving node with the built-in one; everything else remains unchanged.

Batch production is supported; simply change the image folder address.

因看到很多转真人的工作流很复杂，有的还要lora。经测试多个，有的要一堆节点，我还遇到节点管理器坏了，一直假安装，下载zip又缺依赖，实在没辙就自己研究，测试的工作流发现lora没啥用（可能我测试方法不正确），或者说真正起作用的是“参考Latent”这个节点。有的工作流没有控制比值，无法调节重绘幅度。

本人喜欢极简，几乎就是comfyui自带的节点。gguf可以替换成safetensors，保存图像节点换成自带的，其他的不用变。

支持批量生产，改一下图片文件夹地址即可。