Amidst the massive controversies with Stability AI, the company revealed its latest take on its AI image generator, debuting the all-new Stable Diffusion 3, promising enhanced capabilities. The new AI tool is more advanced than its two previous versions, now promising up to 8 billion parameters, currently in early preview that now accepts signups to join its waitlist.
The company promises massive improvements in all aspects of the image generator tool, centering on performance and safety before its official release.
Stable Diffusion 3 from Stability AI Brings Massive Improvements
Stability AI announced the arrival of its Stable Diffusion 3, the latest and most advanced version of its AI image generator available in the market. Calling this the "most capable text-to-image model," Stability revealed that it is now using a diffusion transformer architecture in this program to deliver improved performances in aspects like "multi-subject prompts, image quality, and spelling abilities."
The company also revealed that the Stable Diffusion 3 suit of models range from 800 million to 8 billion parameters, offering significant scalability and quality for creative needs.
Moreover, Stability AI regarded its new focus on safety in this release, as one that would prevent bad actors from the misuse of Stable Diffusion 3 by adding safeguards.
Join the Waitlist for Stable Diffusion 3's Early Preview
For now, Stable Diffusion 3 is in its early preview, and only selected users are part of this experience. However, those who want to join this preview may do so by signing up for Stability's waitlist for the AI.
Through this early preview, Stability would collect insights and reviews regarding their new tool to use for further improvements and upgrades for Stable Diffusion 3 in various aspects.
Stable Diffusion and Deepfake Problems Now
Over the past year, Stability developed its Stable Diffusion to improve more of its outputs, including the fix for the infamous "spaghetti hands" that many observed from its creations.
AI tools in the world have faced significant scrutiny and controversies from various parties, and this is because their capabilities are now manipulated by unknown sinister individuals which create deepfakes. More importantly, the rise of deepfake porn in the country became a massive problem recently, going as far as creating Taylor Swift's AI scandal which is now subject to legal action.
In the many reports behind deepfake porn, Stable Diffusion is among the frequent tools mentioned as the source of these malicious images that hurt the feelings and reputations of its victims.
While Stability AI is facing this controversy, they are still ramping up its plans to deliver new AI experiences for the world, including the improvement of its image generator tool. Stable Diffusion 3 is said to be the most advanced version from the company now, centering on massive improvements like multi-subject prompts and up to 8 billion parameters available in this early preview.
