Guides

Long-form notes on OpenAI service tiers, the 272K long-context cliff, prompt caching and how each lever actually multiplies a bill. Worked examples use the same price table as the calculator.

Tiers

• OpenAI Batch API: how the 50% discount works and when it is the wrong choice Batch is a 0.5× multiplier on the whole request, not a separate price list. A worked enrichment job, the 24-hour window that forbids agent loops, and the arithmetic of trimming first. 5 min read · updated 2 October 2026
• OpenAI Fast mode: what 2x the rate actually buys you Fast is a 2x multiplier for up to 2.5x the speed, bought per request rather than per account. The clean result is that buying Fast for a share of traffic adds roughly that share to the bill. 5 min read · updated 2 October 2026
• Flex vs Batch: both cost half, but they fail differently Two tiers halve the rate and differ only in how they break. Flex is synchronous and may be unavailable; Batch is asynchronous and bounded by 24 hours. A retry strategy for each. 5 min read · updated 2 October 2026
• How to choose an OpenAI service tier for a request Batch and Flex both halve the rate, and they are the largest saving most teams never set. A decision order that maps nine common request types to a tier, a model and an effort level. 5 min read · updated 2 October 2026
• OpenAI Ultrafast: 6x the price, 6x the speed, Astra only The 6x multiplier lands on input, cache read, cache write and output together, and only GPT-6 Astra publishes Ultrafast pricing. It pays only when wall-clock time is the binding constraint. 5 min read · updated 2 October 2026

Caching

• OpenAI prompt caching: what cached input actually costs Cache reads cost 5% to 10% of the input rate and cache writes cost 1.25x, so a reused prefix pays for itself on the second call. The ratio survives the long-context cliff. 5 min read · updated 2 October 2026

Cost

• Eight levers that cut an OpenAI API bill, ranked by how much they save A ranked list of the eight changes that actually move an OpenAI bill, from the half-price tiers down to the regional surcharge, with a quantified effect and a reason to skip each one. 6 min read · updated 2 October 2026

Long context

• The 272K long-context cliff, and how to stay on the right side of it The long-context threshold is a cliff, not a slope: crossing 272,000 input tokens roughly doubles the cost of the entire request. Two worked examples and four ways to stay underneath. 5 min read · updated 2 October 2026