Discover a unique face for your character! This workflow is designed to generate diverse, random faces, based on randomly generated names. By doing this, we sidestep the use of celebrity images and names, ensuring a pool of original names and faces for your characters, free from ethical dilemmas.

How to use it:

1. Generate random names using a name generator. I use https://www.behindthename.com/random/

2. Paste the random names into the prompt list.

2. Pick the SD1.5 model you want to use. I used  https://civitai.com/models/4201?modelVersionId=130072 for the examples

3. Find your character!

Tips:

1. I'll be honest, I don't like or use this workflow. The faces generated from SD1.5 are just... meh. And getting character consistency across seeds is really hard. So if you have the VRAM, I'd recommend face finding on an XL model. Use a Turbo model if it's too slow. But if you want to try your luck... maybe you'll find the perfect face. Who knows? But otherwise I just recommend the SDXL workflow. It's a much better experience in my opinion.

https://openart.ai/workflows/wilburn/face-find-sdxl/OsArpYrUDTAuqYsR0nKc

2. Keep the prompt as simple as possible to keep the focus on the name itself.

3. Don't forget to reset the counter to 0 every time if you want to start at the beginning of your list

4. If you enable "Extra options" under "Queue Prompt" you can run this in batches. I usually put 10 for the batch. This will queue up 10 at a time. Note: this is not the same as the latent "batch_size" which will send multiple images through the sampler at once. The queue batch just sends multiple prompts to the queue, and will auto-increment the list index.

Next Steps (How I use this in my overall workflow):

1. I generate like 100 faces using one of my Face Find workflows (usually the SDXL version)

2. Pick about 5 faces I like, then generate several more faces with the same name

3. Pick the 1 face I like best for the character, upscale and detail the faces using another workflow (though again, SD1.5 is hit-or-miss with this), though sometimes I just use the original image.

4. Use the face image in a face replace workflow, my preference is IPAdapter, but FaceDetailer and Reactor work well too.

Additional comments:

-- the names included in the workflow were generated randomly using all categories on behindthename. I generated 5 with just first name and surname and 5 with first, middle and surname. I didn't cherry-pick them, these were just the 10 random names I got. You'll notice if you use the names I have in the workflow that the last one has Cola as the first name. That came from the randomizer and I thought nothing of it except that soda bottles kept showing up in my faces. Ugh.

-- I cherry picked the seed that turned out the best for this workflow. All the faces shown in examples are the same seed from a single run. But they are from the 4th or 5th run.