I used Claude Fable 5.1 in the video and compared it with ChatGPT and Gemini on the same topic.
In conclusion, I thought it was pretty good. However, I think it's best to only ask for occasional tasks. I think Opus 5 is usually enough. The reason is simple: it's extremely slow.
This is wild! Way beyond expectations!
AnthropicAI's Fable 5.1 just one shotted about 90% of the build ↓ I gave it a rough proof-of-concept I build with Opus 5 yesterday, a messy voice memo + Figma URL's to the design. 17 minutes later it created this (video below)👇🏻 It
Fable 5.1 is out 🧡
Same price as Fable 5, but cache reads cost a quarter 25% cheaper on normal tasks, up to 45% on agentic ones science benchmark 24.7% → 52.6% for most work, Opus 5 still does the job
The anatomy of a second brain that writes itself
One Obsidian vault, five folders, and a loop that files every good answer back into a wiki you never touch the structure: CLAUDE.md + AGENTS.md - the constitution, same rules whether Claude or Codex is running the loop raw/ -
You have reached the end of the archive
All of Claude Opus 5