A 27B model one-shot recreates 618 visual styles as SVG artworks, text-only
Seromelhor · reddit · 2026-09-22
Developer Seromelhor tested whether Qwen 3.8 27B truly understands visual style: in one-shot, text-to-text runs (no vision, no recursion), the model recreated the visual language of 618 designers, artists and movements as new SVG artworks and pages, taking 1.5 days. SVG was easier and more creative; pages needed heavy prompt work to escape clichéd "AI-generated" style. Naming artists directly increases component design variety. All 618 examples are at artificialguy.com/svg-museum.
More from Models
- Grok 4.7 launches with notable gains at same price and speed — elonmusk · 2026-09-22
- rauchg: Grok 4.7 nailed a hard binary reverse-engineering task, and fast — elonmusk · 2026-09-22
- Paradigm's Limite model skips heavy instruction tuning to maximize math per parameter — tensorqt · 2026-09-22
- Unverified: Grok 4.7 reportedly out with notable gains at same price and speed — petergyang · 2026-09-22
- Multi-turn evaluation changes what it means to pass the Turing test — CShorten30 · 2026-09-22
- Grok 4.7 launches at same price and speed as Box demo shows it catching an $82K duplicate invoice — elonmusk · 2026-09-22