Source: Compaction (OpenAI guide)

OpenAI's guide to compacting long Responses API conversations, server-side via context_management or explicitly via /responses/compact, using opaque encrypted compaction items.

Contents
  • What: official API documentation page
  • Publisher: OpenAI
  • Date: living document; read 2026-11-14
  • Original: Compaction

Summary

Two ways to shrink a long conversation so it keeps fitting the context window, both producing an opaque compaction item the model can continue from.

Key claims

  • Server-side: context_management: [{type: "compaction", compact_threshold: N}]; when the rendered prompt crosses N tokens, the server compacts, emits a compaction item in the same stream and prunes context.
  • Standalone: POST /responses/compact takes a full window and returns a new, smaller one; it is stateless.
  • The compaction item is encrypted and not human-readable; compacted windows may also keep some earlier items.
  • With manual input chaining, items before the latest compaction item can be dropped; with previous_response_id chaining, do not prune manually; never prune /responses/compact output.
  • Both modes work with ZDR when store=false.

Feeds

OpenAI Responses API

Suggest changes →