Use case
GPT-5.2 prompt caching
prompt caching notes for gpt-5.2: 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 $0.910 on corouter (input $0.350, output $0.560).
Cache-read rate
Cache read for gpt-5.2 lists at $0.175 and is $0.035 on corouter per million tokens. Uncached input is $0.350.
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-5.2 shares the 272K window and list rate with the Codex IDs but does not carry the Code tag. Send gpt-5.2 for chat, reasoning, and vision at that rate.
Questions
- Which model ID should the request send?
- Send gpt-5.2. 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.