Stability AI, the company behind Stable Diffusion, the popular open source machine learning model that generates images from users’ text descriptions, is announcing the arrival of StableStudio as the open source version of DreamStudio that will allow progress in its development.
They explain that with the initial release of Stable Diffusion, DreamStudio has served as the primary interface for regular users to access the models and features they have added over time, with SDXL being the most recent rendering model to come to DreamStudio. .
Leaning back on the open source developer community
They seek that by generating a community of developers on StableStudio, more progress can be made in the development of DreamStudio than by focusing development directly on DreamStudio as a closed source product, an approach that is not new in the company, since it has been relying on code for a long time open to boost the development of its products.
More explicitly they state that:
Our goal is to work with the broader community to create a world-class user interface for generative AI that is fully controlled by users. Although DreamStudio will remain Stability’s hosted implementation of StableStudio, our goal is to foster a project that can surpass anything developed by a single company.
Open source as a promotion of transparency and trust
Added to this is what the founder and executive director of Stability AI, Emad Mostaque, recently pointed out to the press via Zoom, pointing out that having AI tools in open source makes it possible to increase public confidence, considering that “the open models will be essential for private data”
The truth is that, as they comment in The Verge, StableStudio will be together with DreamStudio and will potentially compete with it, so that whoever wishes can make their own version even with the modifications they deem appropriate.
In this regard, the company recently came to consider earning revenue from offering custom versions of DreamStudio to its customers, something that is made less viable with today’s announcement.
But now his strategy is different:
Our goal of creating a great multi-modal experience for generative AI has endured. To that end, we’ll be working on releasing our next chat interface within the StableStudio project.
With a list of aspects pending improvement
The arrival of StableStudio has meant moving its active development to the public repository and they plan to launch rewards for improvements and new functions, aiming to improve aspects such as:
– Local inference via WebGPU
– Local inference via stable broadcast webui
– Desktop installation
– ControlNet Tools
Those interested in participating will have all the information in the README file included in the repository.
More Information/Image Credit: Stability AI