Stability AI, the generative AI company that gained recognition for its open-source image generation tool, Stable Diffusion, has announced the beta release of its latest image generation model, Stable Diffusion XL (SDXL). The model offers more advanced features and photorealism capabilities than its predecessor, Stable Diffusion 2.1.
The Stable Diffusion XL beta can be accessed via Stability’s API and DreamStudio programming suite for enterprise developers. In addition to enhanced image synthesis and face generation, SDXL beta enables short prompts to generate detailed and descriptive images. The model also supports image-to-image prompting, inpainting, and painting, which seamlessly extends and builds upon existing images.
Stability AI, based in London, has experienced remarkable success since its establishment in 2019. However, the company has faced pushback from artists who object to their work being used as training data for Stable Diffusion models.
Despite the objections, Stability AI has continued to evolve and improve its AI art models. However, in January, artists filed a copyright infringement class action lawsuit against Stability AI and Midjourney over AI art.
As generative AI advances, the concerns of artists and others will likely continue to be raised. Nonetheless, companies like Stability AI are helping to drive innovation and transform several industries with their cutting-edge AI models.