stable diffusion artist collaborations

Astronaut Horse 🐎

Astronaut horse is a series of open-ended collaborations between visual artists / students and a group of professional software engineers and machine learning researchers who began working with Stable Diffusion together at the Recurse Center in late 2022.

The name astronaut horse refers to the common AI-generated image of an astronaut riding a horse that often acts as a benchmark for the capabilities of generative image models.

Stable Diffusion ⚙️

Stable Diffusion is a free and open-source alternative to image-generating AI tools like DALL-E and Midjourney which remain private and for-profit. The tools developed for this project are similarly intended to be free for artists to use and help develop.

wikipedia link

Artist Collaborations 🎨

Collaborations with visual artists began in 2022 with a series of Stable Diffusion fine-tunings (using textual inversion training) on artist-chosen images of their own artwork as well as images that inspired their studio practices. Collaborations have since expanded to be open-ended explorations seeking new ways to incorporate generative models into artist's long-term creative practices.

Community Grant 🤗

Since late 2022 this project has been generously supported by a Community GPU Grant from Hugging Face, one of the most prominent groups providing free resources to the larger Stable Diffusion community. This grant allows the GPU-powered app linked above to stay running and accessible to anyone any time.

Workshops & Tools 📝

The project has several student / community workshops planned focused on hands-on creative projects introducing participants to basic coding skills and the conceptual foundations of generative AI tools. The resources for these workshops are freely available on this website.

workshop links

Upcoming Things 📅

Contact / Discord 🗣️

If you'd like to see more images, get more information or if you're interested in potentially collaborating as an artist / researcher / engineer please feel free to reach out directly. Just describe your interest in the project and include some links to your recent projects / personal site / github / linkedin etc.