Stable Diffusion 2.0 and the Importance of Negative Prompts for Good Results

As an unexpected surprise, StabilityAI released Stable Diffusion 2.0 last week, the next major version of the text-to-image AI that has been warping the entire ecosystem. Architecture-wise it’s mostly the same, except with a new text encoder (OpenCLIP instead of OpenAI’s CLIPText). StabilityAI boasts that Stable Diffusion 2.0 has better performance quantitatively, but art in the end is subjective.

Within 24 hours after release, users on Reddit and Twitter noted that the new model performed worse than …

diffusion good importance negative stable diffusion