Astral Codex Ten

Stable Diffusion

3 min read

Stable Diffusion

Article

Stable Diffusion is a recurring brand in the Astral Codex Ten archive, appearing 2 times across 2 issues between August 30, 2023 and July 18, 2025. The archive places it in contexts such as “to get my desires across to this stable diffusion model”; “not the kind generated by stable diffusion”. It most often appears alongside ACX, Scott, 16th century Spain.

Metadata

  • Category: Brands
  • Mention count: 2
  • Issue count: 2
  • First seen: August 30, 2023
  • Last seen: July 18, 2025

Appears In

Source Context

Recovered passages from the original issue text. When the raw archive preserved outbound links inside the source passage, they are listed directly under the quote.

Highlights From The Comments On Fetishes
August 30, 2023 · Original source
But the rest was really good — consistent characters and outfits, just missing the bottom. After a few hours of experimentation and re-rolling we could generally end up with the ideal full-body image, including feet. I developed a whole toolbox of tricks, from changing the aspect ratio to be really tall to blending in an example pose as a second image. One of the tricks to get this was to get really specific about the feet. What kind of shoes and socks? Mentioning feet a few times with emphasis. Things like that. My language was one that I imagine a foot-fetishist’s google search history. "Tall woman with feet showing, high heels, extra feet please with an order of foot." Or extreme detail like "orange nikes with shoelaces and socks and shadows under the shoes" That would generally convince the AI that the feet should also be shown. Not all the time, but some combination helped. I dreamt of feet, and kept trying to come up with more synonyms to get my desires across to this stable diffusion model.
Your Review: Islamic Geometric Patterns In The Metropolitan Museum Of Art
July 18, 2025 · Original source
To the extent that I’ve been able to create or replicate some of these patterns, the process reminds me of procedurally generated art — not the kind generated by stable diffusion, but the old school kind, generated by explicitly specified algorithms, autonomous systems, or the manipulation of randomness. As I understand it, there is often an experimental component: the artist begins with a vision, cooks up procedures, sees what comes out, then iterates, refining the method or sampling new random outputs with different seeds.