OpenAI Responses API adds 2h and 8h background completion windows at 30% and 40% off (public beta, November 17, 2026)
Set background_options.completion_window to 2h or 8h on background Responses calls to cut token rates by 30% or 40%.
Your keys
0 of 1 ready
Create each key on its own page, paste it below, then copy the finished lines into your project's .env file. Your keys stay in this tab. They are never sent to Wirestack or saved; we only count that the panel was used.
- OpenAI API keyOPENAI_API_KEYGet key
Click Create new secret key
What changed
On November 17, 2026 OpenAI added discounted completion windows for background requests in the Responses API, as a public beta. They work on gpt-6.2-astra, gpt-6.2-sol, gpt-6.1-terra and gpt-6-luna.
You send background: true and set background_options.completion_window:
| Window | Model token rate |
|---|---|
soonest (default) | Standard |
2h | 30% below Standard |
8h | 40% below Standard |
The discount applies to every eligible request, follow-ups included. Rules that affect your code:
- Leave
streamunset orfalse. - The windows need Standard processing. Set
service_tiertodefault. Withauto, a request gets the discount only if the API picks Standard. 2hand8hrequests share the Batch/Background rate limit with Batch API jobs, and input tokens count when you submit.soonestrequests keep the regular limits.- Standard requests cannot reuse prompt cache entries that
2hor8hrequests created, even for an identical prompt. - For a multi-step job, set
background_options.continuation_idto an ID your app picks. Follow-ups that reuse it with the same model, service tier, background setting, window and project can get higher queue priority. Leave it unset for a single response.
You get results by polling GET /v1/responses/{response_id} until the status leaves queued or in_progress, or through the response.completed webhook. Deduplicate webhook deliveries with the webhook-id header and check that event.data.id matches the response ID you saved.
The Python SDK openai 3.53.0 and the Node SDK openai 7.58.0, both released November 17, 2026, add the 2-hour and 8-hour Responses background windows.
