Reading up on Sonnet 4.5
2 deep · digging since dec 17, 25
- What Claude Code chooses
A report finds Opus 4.6 strongly recommends Vercel, Tailwind, shadcn/ui, and GitHub Actions, while avoiding Redis and React as default choices.
- Prompt caching: 10x cheaper LLM tokens, but how?
Prompt caching cuts LLM token costs by 10x and latency by up to 85% by caching attention key-value projections for repeated prompt prefixes.