Canadian cosplayer and influencer. For best resemblance use together with the pair embedding (LINK) and adjust weights as needed.
Description
FAQ
Comments (3)
Looks promising, but could we get a hint of the prompts used for the samples? I've been trying to use the lora and TI together at 0.9 as recommended and the face keeps coming out pretty ugly.
I'm using basically the same prompt all model sample images. It goes like this: "RAW portrait photo, (embedding:MGI2:0.9) in busy Vancouver street, embedding:tech\fcDetailPortrait" This is ComfyUI syntax because images are made with ComfyUI (that's why you can't see prompts, Civitai simply is not able to parse and display them as with A1111 prompts). But they are there, you can copy the 10 Nodes button near the Workflow row of data and paste it either to Comfy canvas, or (if you don't have it) just to Notepad, you will see the prompt at the end of that pasted data.
@dolirama126 Oh cool, will you look at that yes I see what you mean looking at the pics in Notepad++. Thanks!