Astronaut Horse: Stable Diffusion Artist Collaborations
summary
Open-source Stable Diffusion workflows get specific: a pipeline uses textual inversion trained on 5–15 images to create new tokens, batch-prompting produces hundreds to thousands of outputs, and deployments run on a Hugging Face Community GPU grant with Gradio + Spaces auto-reloading concepts every 24 hours. Built tools include colab notebooks for training, a batch upscaler for gallery prints, a compound-token “artbot” tab, and weeklong workshops leading to two physical exhibitions that successfully hang and print the work.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 7:03Project origins
- 13:15Stable Diffusion basics
- 20:43Live web app
- 25:18Artist collaborations
- 33:17First exhibition
- 36:49Studio practice
- 44:17Fine art with diffusion
- 47:50Q&A
related talks
29mIntegrating Stable Diffusion Generative AI in Artist Workflows
10mDiffusion models
28mBeyond DALL-E: Advanced Image Generation Workflows with ComfyUI
28mComputing Large Foundational Models
47mFuzzy Humanity Put into Production
01h 27mImage Generation with AI
25mDecoding Algorithmic Artistry: Visual Generative AI
46mHow to Make Stable Diffusion Generate Not Just Anime and Cats, but You