RT @pharmapsychotic: #stablediffusion2 uses the OpenCLIP ViT-H model trained on the LAION dataset so it knows different things than the Ope…


#stablediffusion is just three months old. Three. Yet, the criticism is ferocious.
If the same people would analyze and criticize with the same ardor the v1 of software released 3 months after its GA, no product would survive.


RT @minimaxir: It’s appearing that the key to use Stable Diffusion 2.0 will be that you will actually have to use negative prompts, as they…


RT @lopp: Midjourney AI engineers are predicting an order of magnitude improvement in image generation results every 2 months for at least…


RT @fchollet: Many people in tech have so little exposure to philosophy (or the humanities in general) that when they get exposed to old id…


While @StabilityAI was busy releasing #StableDiffusion2, @drawthingsapp was busy bringing Stable Diffusion 1.5 (and many other models based on it) to the iPad.

Lots to try. https://t.co/a2rBLpmlvy


Humans rediscover the lost art of articulated narrative (probably lost after Mervyn Peake finished writing Titus Groan) because of Stable Diffusion.

Titus Groan becomes a mandatory textbook in Prompt Engineering classes. https://t.co/7pMm4odZ6Z


RT @EMostaque: Some user notes on V2! 🧵

V2 prompts different and will take a while for folk to get used to.

V2 is trained on two models,…


Stable Diffusion 2.0 Release — https://t.co/EZ6LA12pT3 https://t.co/pKTnDNtyUq https://t.co/s35adq76jF