Use case
GPT-5.5 prompt caching
prompt caching notes for gpt-5.5: 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 $2.20 on corouter (input $1.00, output $1.20).
Cache-read rate
Cache read for gpt-5.5 lists at $0.500 and is $0.100 on corouter per million tokens. Uncached input is $1.00.
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-5.5 keeps the high GPT-5.6 Sol list rate and the Code tag. Send gpt-5.5 when a workload was already qualified on that exact ID.
Questions
- Which model ID should the request send?
- Send gpt-5.5. 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.