We present LAION-COCO, the world’s largest dataset of 600M AI generated high-quality captions for publicly available web-images laion.ai/blog/laion-coc…

I translated 3B captions from hundred of languages using M2M100 model and we released them as laion-translated LAION. Learn more at: laion.ai/blog/laion-tra…
Here's some examples 👇

imagen video: high definition video generation with diffusion models

paper: imagen.research.google/video/paper.pdf
blog: imagen.research.google/video/

Recording is now live with Stability AI 's Emad interview with hosts Claire Silver 🌸 & Ben Roy

One of the most inspiring & thought provoking talks 💫

Thank you for all you're doing to make the world a better place Emad 💜

twitter.com/i/spaces/1YpJk…

make-a-video: text-to-video generation without text-video data
paper: makeavideo.studio/Make-A-Video.p…
project page: makeavideo.studio

CLIP guided #stablediffusion with the newly released LAION CLIP models using 🧨diffusers

GitHub: github.com/huggingface/di…
Colab: colab.research.google.com/github/hugging…
CLIP models: huggingface.co/models?search=…

Guiding Stable Diffusion with our CLIP H: 'Professional HDR photo of a polar bear holding a lollipop on a rooftop in Hong Kong looking up at a UFO in the night sky. A UFO flies above the polar bear. The polar bear holds a lollipop on a rooftop. The background shows Hong Kong.'

Inspired by various recent efforts to make sense of the text2img datasets - here's all 12M captions from LAION-Aesthetics with score > 6, embedded with CLIP and UMAP'ed to 2d. Color is the domain of the image URL.

laion.ai/blog/laion-aes…

#StableDiffusion a beautiful photo of a plushy Darth Vader Teddy Bear, trending on Instagram

#StableDiffusion a beautiful photo of a plushy Deadpool Teddy Bear, trending on Instagram