Use case
GPT-5.6 Luna prompt caching
prompt caching notes for gpt-5.6-luna: the rate card, a worked volume, and when to send this ID. A sample of 1 million input tokens plus 200 thousand output tokens is about $0.088 on corouter (input $0.040, output $0.048).
Cache-read rate
Cache read for gpt-5.6-luna lists at $0.020 and is $0.0040 on corouter per million tokens. Uncached input is $0.040.
Only the input that actually hits cache is billed at the cache-read rate. This page does not promise your prompt will hit.
Where this ID sits
GPT-5.6 Luna is the low GPT-5.6 rate, chat and vision only. Send gpt-5.6-luna when volume matters more than a reasoning or code tag.
Questions
- Which model ID should the request send?
- Send gpt-5.6-luna. Do not put a different name in the model field.
- What is the minimum top-up?
- Prepaid credits, $5 minimum top-up, no subscription tier.