Tag
7 posts tagged.
A thinking block stays valid against the bytes you sent. The prompt cache keys on the prompt the model renders. The rows where those two disagree cost money with no error.
Anthropic's batch docs tell you to swap the 5-minute prompt cache for the 1-hour one. The swap costs a 60 percent heavier write on every miss, and the hit rate it has to reach to pay for itself is 57.6 percent at the bottom of Anthropic's own published band and outside that band at the top. A correction to my own September post.
An Anthropic Priority Tier commitment is drawn down alongside your ordinary rate limit, not on top of it. Reading the service tiers page next to the pricing page and the models list turns up a 20x burndown spread, a single cache read rate that only holds because of the exclusion list, and three of four current models excluded.
Context engineering on the Claude platform is not a setting. It is four mechanisms at four points in the pipeline, and the one everybody enables first pays for context window with prompt cache.
The Batch API's default prompt cache expires in 5 minutes while the batch itself runs for up to an hour, so the two cost levers quietly cancel each other out. Six documented constraints, and the single assumption underneath all of them. Field log, September 2026.
Claude Code auto compact fires on a per-model token boundary, not a percentage. What it costs is set by prompt cache warmth rather than context size, and a path-scoped rule does not come back afterwards until Claude reads a matching file again.
Claude Code's /fast is not a speed dial, it is a mode. The first toggle in a conversation reprices your whole context, the toggle moves your model one way and does not move it back, and asking for fast is not the same as getting it.
