OpenAI Introduces Asynchronous Background Completion Windows for Responses API
In short: OpenAI has launched a public beta for the Responses API that allows you to select specific background processing windows. By opting for a two-hour or eight-hour completion window, your company can achieve substantial cost savings on token usage.
This summary was generated automatically by AI from OpenAI API Changelog's publication. It is our own text, not a copy of the original — facts, figures and quotes belong to the source, linked above and below.
What changed?
- 1New two-hour and eight-hour background completion windows in public beta.
- 2Token rates for the two-hour window are 30% cheaper than the Standard rate.
- 3Token rates for the eight-hour window are 40% cheaper than the Standard rate.
Why it matters
This update is ideal for high-volume, non-urgent tasks like log analysis, automated documentation updates, or background data processing, where immediate model response is not required but cost efficiency is a priority.
What it means for AI agents and contact centers
If you manage high-volume asynchronous tasks, this allows you to optimize operating margins. Compare your current standard batch processing latency against these time-bound windows to see where you can switch to cheaper throughput without impacting system reliability.
🧪 Worth Testing
Testing these windows helps determine if the delayed response time is acceptable for your non-realtime automated workflows, potentially significantly reducing API expenditure.
Sources
- OpenAI API ChangelogOfficialPrimary sourceOriginal article →„Choose a two-hour or eight-hour background completion window in the Responses API public beta. Model token rates are 30% lower than Standard for the two-hour window and 40% lower for the eight-hour wi“17 Nov 2026, 02:00
- Published by source
- 17 Nov 2026, 02:00
- Found by our system
- 17 Nov 2026, 20:14
- Summary generated
- 17 Nov 2026, 22:15
This article was written by AI from the original source. Facts, numbers and prices come from the source; missing values are marked “Not specified”. Legal notice, copyright and privacy