laion-ai / erlich

Generate a logo using text.

  • Public
  • 348.3K runs
  • GitHub
  • License

erlich is the text2image latent diffusion model from CompVis (with additions from glid-3-xl) finetuned on a dataset collected from LAION-5B named Large Logo Dataset. It consists of roughly 100K images of logos with captions generated via BLIP using aggressive re-ranking.

For more info see the README.md for ldm-finetune.