If you build on the Claude API, this changelog is the authoritative source for the behavioral shifts that will affect your code: thinking enabled by default with adjustable effort levels, tool definitions that can change mid-conversation for long-horizon agents, and breaking changes to max_tokenThe chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.Full definition → accounting and thinking opt-out.
Terms in this piece · Glossary
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
