search

openai / glide-text2im

GLIDE: a diffusion-based text-conditional image synthesis model
MIT License
3.53k stars 500 forks source link

some questions of the result picture #11

Closed lixiangMindSpore closed 2 years ago

lixiangMindSpore commented 2 years ago

2 prompt = "a woman"

What's wrong with this?

lixiangMindSpore commented 2 years ago

I use this image

woctezuma commented 2 years ago

prompt = "a woman"

What's wrong with this?

See #2. Click on the link to my repository, then read the model limitations in the section called Safety Considerations.

unixpickle commented 2 years ago

As pointed out by @woctezuma in the above link, the full model has not been released. The model in this repository was trained on a filtered dataset containing no images of people, and it is 10x smaller.