eval — A repeatable test for AI quality — a set of tasks plus scoring — used the way software teams use test suites, because model output is too variable to judge by eyeballing.
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
context compaction — Summarizing an agent's earlier conversation to free room in the context window so a long session can keep going.
Why it matters
If you run agents over long sessions, context compactionSummarizing an agent's earlier conversation to free room in the context window so a long session can keep going.Full definition → strategy silently determines what your AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → forgets — this lays out a repeatable way to measure which summarization approach preserves task-relevant context windowThe maximum amount of text a model can consider at once — its working memory for the current conversation or task.Full definition →, with side-by-side results against OpenAI and Anthropic's compaction behavior. Read the methodology as the takeaway and treat the winner claim as vendor-flavored.
