Thread by @EMostaque

Delighted to announce the public open source release of #StableDiffusion!

Please see our release post and retweet! stability.ai/blog/stable-di…

Proud of everyone involved in releasing this tech that is the first of a series of models to activate the creative potential of humanity

We are also delighted to announce the public beta release of #DreamStudio, our reference implementation to allow anyone to create anything.

The public API will be available here shortly to extend your creativity and enable new experiences.

beta.dreamstudio.ai

We have more to come and expect an explosion of originality in UI/UX and new ways to create both from our efforts and those of the community, which everyone should join at discord.gg/stablediffusion !

Developers, creatives and those inspired by this, all are welcome to come & dream

A sneak peek at some of the futures to come for #DreamStudio, including local GPU usage, animation, logic-based pipeline flow with more such as inpainting, dynamic variation and latent space exploration to come..

Very proud of the whole team and look forward to the future

We look forward to highlighting the work of all those who build on #StableDiffusion (note per the license you have to say if you are! But use the hashtag anyway as that's the cool thing to do), as well as supporting them in various ways with some big announcements to come..

To the license itself and the extra work done, this comes in two parts.

The license is ethical use and should be distributed with model and made available to end users. It is permissive and respectful of individual responsibility.

We hope it sets a standard for these models.

The safety filters incorporated in the default package can be adjusted (you can even add things like removing clowns), but are important given the various limitations highlighted in the model card for this general first model that is reflective of humanity huggingface.co/CompVis/stable…

... or at least what humanity is reflected in a scrape of the internet.

It does not target any specific group and sets a basis from which far better and diverse models can be built.

We will be building models and datasets for all cultures, races and countries.

Plus...

We are making it so every individual can create, fine-tune and train their own models that take their information and condense them down to a few gigabytes.

These are really generative search engines that take our information and turn it into compressed knowledge.

By making these frameworks we look forward to the diversity this will create and encourage you to come and share on our community and let us know how we can support you.

Many more models to come but this is only possible through the work of our communities, partners and you.

Today we all made the World a more creative, happier and communicative place.

More to come in the next few days but I for one can't wait to see what you all create.

Let's activate humanity's potential.