Use case
GPT-5.6 Terra prompt caching
prompt caching notes for gpt-5.6-terra: 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 $0.880 on corouter (input $0.400, output $0.480).
Cache-read rate
Cache read for gpt-5.6-terra lists at $0.200 and is $0.040 on corouter per million tokens. Uncached input is $0.400.
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 Terra sits between Sol and Luna on the GPT-5.6 rate card and does not carry the Code modality tag. Send gpt-5.6-terra for long-context chat and vision at the mid rate.
Questions
- Which model ID should the request send?
- Send gpt-5.6-terra. 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.