//=time() ?>
ruDALL-E Aspect Ratio Text-to-Image from @AVShonenkov is now included in Visions of Chaos.
https://t.co/YJcHEjBc73
"a black and white photo of a werewolf"
"a lounge room"
"an evil clown"
"dense woodland"
The reaction of artists to Dall-e-2 reminded me of this scene from watchmen.
On the wait list for Dall-E 2
Anyone know a way to get access now?
It will change the approach to making images so profoundly, I can't wait to get my grubby paws on it! 🐩🐩
DALL-E2 es un nuevo sistema abierto de Inteligencia Aritificial que permite crear imagenes a partir de descripciones, combinando sujetos, objetos, temas y estilos.
Por ejemplo:
Astronautas jugando basket con gatitos, en estilo libro de niñes.
Here's what a few years of progress on text-to-image generation looks like, one prompt at a time.
"Frank Sinatra as a purple alien in surrealist style"
1) AttnGAN (2018)
2) CLIP+VQGAN (2020)
3) CLIP+Diffusion (2021)
4) DALL-E 2 (2022)
DALL-E illustrations of "all roads lead to the same place, peaceful girl meditating for eternity"
Trying out a new avi generated by DALL-E, I like that it captures my vibe from an old pic
DALL-E illustrations of "fantasy illustration of cyclical scientific discovery throughout history, hopeful and bright" and "independent scientists joining back into the tree of knowledge, cyclical nature of life, bright pastel"
DALL-E's art is incredible. Here we have "illustration of independent researchers walking above ground out of mycelium" and "pastel hopeful and dreamy illustration of a girl blooming out of a lotus flower"
@liminal_warmth @nickcammarata This is what @midjourney came up with. Would love to see what DALL-E 2 would create but, alas, I don't have access yet.
I would like to nominate "an astronaut playing basketball in space with cats" as the top DALL-E-2 prompt thus far #openai
An astronaut playing basketball with cats in space (as a children's book illustration | in a minimalist style | in a watercolor style | )
Inspired by @OpenAI 's DALL-E 2 post's prompts...
Try it out in @gradio on @huggingface Spaces: https://t.co/VOANiDrvdF
While I wait for DALL-E 2 I'm playing around with latent diffusion model LAION-400M. If you ask it to make your favourite things it'll show you very loose interpretations of them but at least it's very clear about telling you what they are