Anthropic describes Haiku 5.5, announced October 7, 2026, as its fastest and most efficient ModelA computational system that uses learned patterns to produce an output from information supplied to it.Full definition → in the Claude 5.5 family, built for high-volume, cost-sensitive work such as summarization, classification and routing. Its Haiku product page also says it is the first Haiku with effort controls, which let teams tune cost against intelligence for each task.
PromptInformation or instructions supplied to a model; in a chat, the request you type is often called your prompt.Full definition → length changes the price substantially. For Claude Platform use, the same page lists input/output rates of $0.10/$0.50 per million TokenA piece of text, such as a word, part of a word or a punctuation mark, that a language model represents as numbers and handles as a unit.Full definition → for prompts up to 100,000 tokens, versus $0.50/$2.50 above that threshold. Budgeting therefore requires knowing the size of the material sent to the model, as well as how often it runs.
Terms in this piece · Glossary
Model — A computational system that uses learned patterns to produce an output from information supplied to it.
Prompt — Information or instructions supplied to a model; in a chat, the request you type is often called your prompt.
Token — A piece of text, such as a word, part of a word or a punctuation mark, that a language model represents as numbers and handles as a unit.