ELLA icon indicating copy to clipboard operation
ELLA copied to clipboard

Run Astral Codex Ten's prompts

Open bakkot opened this issue 1 year ago • 0 comments

There's been occasional attempts to pass AstralCodexTen's image generation challenge (from 2022!), but no one is quite there yet.

  1. A stained glass picture of a woman in a library with a raven on her shoulder with a key in its mouth
  2. An oil painting of a man in a factory looking at a cat wearing a top hat
  3. A digital art picture of a child riding a llama with a bell on its tail through a desert
  4. A 3D render of an astronaut in space holding a fox wearing lipstick
  5. Pixel art of a farmer in a cathedral holding a red basketball

For each prompt, we generate 10 images. If at least one of the ten images has the scene perfectly correct on at least 3 of the 5 prompts, then ACT wins his bet.

DALL·E 3 can get some of these, but it struggles especially to put the bell on the llama's tail, and the key in the raven's mouth.

I would be interested to know if ELLA can get there. Passing this challenge would also be a good way to bring this work to more people's attention - ACX is a pretty widely read blog.

bakkot avatar Mar 12 '24 23:03 bakkot