Every's Fable 5.2 tests: max effort hurts design quality; model built an app overnight unattended
every · x · 2026-11-10
Every's hands-on vibe check of Anthropic's Fable 5.2 finds that cranking the effort setting up often makes design work worse: blind-rated tests showed "high" beat "max" for static design taste, while max helped motion and interaction detail. Team starting points: code at low, research/writing at medium, design at high, animation/interaction at max. The article also highlights a landmark demo—Kieran Klaassen left the model building a video editor unattended and returned to a production-grade app (Everycut) combining Premiere, After Effects, and Ableton-style capabilities within days, calling it the first model that lets professionals build daily-use production tools without reading the code.
More from coding & agent
- Medium launches TK, a writing app that plugs your notes into AI agents via MCP — LuoSays · 2026-11-10
- Developer dedicates a Mac mini as a home for his AI agents — ow · 2026-11-10
- Hyo adds Agent Dock to Obsidian: pin and monitor all your AI chats in one sidebar — evielync · 2026-11-10
- GPT 6.1 agent designs a cascadable broadband half-adder in a 23-minute run — CatAstro_Piyush · 2026-11-10
- Claude Code's PM #2 shares how to do devtools PM in the AI era — EricBuess · 2026-11-10
- Dev demos AI that edits documents live while you think out loud — jungofthewon · 2026-11-10