- Who it helps
- Web developers and people studying small diffusion models.
- Availability
- MIT-licensed source is public; the hosted demo could not be read.
What the model produces
The creator describes twelve deterministic denoising steps using small networks trained on ten hand-drawn sprites. This is a constrained pixel-art experiment.
Where computation happens
The exported page expresses model arithmetic in CSS properties. Training and export use Python; the claimed script-free behavior concerns the generated page.
What you can do
- Read the original method and its limitations before planning your own evaluation. Open reference ↗
Where the evidence stops
No browser run or output inspection performed. Author timings and compatibility remain untested here.
Sources read. Sources read November 9, 2026 UTC. No third-party code, model inference or complete workflow was executed.
Sources & further reading
GitHub · implementation
diffusion.css implementation ↗
Small deterministic sprite generator described; no browser execution.
Metrix187 · page checked 2026-11-09
0 stars · observed 2026-11-09 06:12 UTC
Web · demonstration
diffusion.css demonstration link ↗
Show HN link prompted the stylesheet question. Demo access failed; repository documentation supplied verification.
Metrix187 · search result checked 2026-11-09
More sources reviewed (1)
Hacker News · discussion
Show HN new submissions ↗
Launch titles inspected; individual discussion access was limited.
Hacker News · page checked 2026-11-09
Coverage note: Independent execution and qualifying cross-platform attention remain unverified.
The record over time (1)
Dated source checks and published explanations. An observation date can differ from the original release date.
- · Added to Current
diffusion.css uses CSS arithmetic to generate 16-by-16 sprites.