ml6 / julius

Generate a collection of logos based on your text input. Use longer and more detailed inputs for better results. The first time it takes a few minutes to load the model. Subsequent generations are much faster.

  • Public
  • 4.3K runs
  • License



Run time and cost

This model runs on Nvidia T4 GPU hardware. Predictions typically complete within 7 minutes. The predict time for this model varies significantly based on the inputs.


Generate an ordered collection of logos based on your text input …

Model description

This model was created by a team of engineers at ML6 ( based on minDALL-E ( which combines a VQGAN and a GPT-like transformer of 1.3b parameters. The version implemented here generates three batches of forty images which are then sorted using OpenAI CLIP.

For more details on the model, check out this blogpost

Intended use

Generate ideas for logos, inspire aspiring designers. …

Ethical considerations

The model could generate images that are considered offensive by some people. …

Caveats and recommendations

The model produces the best results when given longer, more detailed text prompts …