Celebrating one year(ish) of Stable Diffusion … and what a year it’s been! 

On August 22, 2022, Stability AI co-released Stable Diffusion alongside talented researchers from LMU Munich and Runway ML. It was a watershed moment in AI that unlocked the power of text-to-image models. One year later, we’ve grown our team by a factor of four, launched more than twenty new products, and become the only independent, …

Introducing Stable LM 3B: Bringing Sustainable, High-Performance Language Models to Smart Devices 

Today, we proudly launch an experimental version of Stable LM 3B, the latest in our suite of high-performance generative AI solutions. At 3 billion parameters (vs. the 7 to 70 billion parameters typically used by the industry), Stable LM 3B is a compact language model designed to operate on portable digital devices like handhelds and …

Ever wondered what those cryptic sampler names like “DPM++ 2s a Karras” actually mean? Look no further.

I was asked to make a top-level post of my comment in a recent thread about samplers, so here it goes. I had been meaning to write up an up-to-date explanation of the sampler names because you really have to dig to learn all of this, as I’ve found out. Any corrections or clarifications welcome! …