Intended to generate similar styled character portraits as to what is shown, because they're embeddings they aren't always 1-1, and are not "TRAINED" as usual I'm using embedding merge. (Which means it's words - vectors - tokens).

If you'd like MORE of these things please see our lazy link list:

Discord: https://discord.gg/5t2kYxt7An

Our website: https://end-media.org/ (This is under heavy re-construction whoops)

Patreon: https://www.patreon.com/earthndusk

Subreddit: https://www.reddit.com/r/earthndusk/

Youtube: https://www.youtube.com/channel/UCk7MGP7nrJz5awBSP75xmVw

TikTok: https://www.tiktok.com/@duskfallcrew

Twitch: https://twitch.tv/duskfallcrew

Instagram: https://instagram.com/duskfallcrew

Membership / Ko-Fi: https://ko-fi.com/duskfallcrew/

Buy Me A Pizza/Coffee: https://www.buymeacoffee.com/duskfallxcrew

Huggingface: https://huggingface.co/Duskfallcrew

Huggingface: https://huggingface.co/EarthnDusk