How to cut your OpenAI API bill

“Prompt caching is enabled by default for supported OpenAI models.”

“Both implicit and explicit caching are supported, where explicit caching gives you more control over which context is written to cache.”

“Only implicit caching is supported.”

“Use `prompt_cache_options.ttl` to control the minimum cache lifetime. The only supported value, `30m`, is also the default. A cached prefix remains eligible for reuse for 30 minutes after its most recent write or reuse, though OpenAI may retain it longer.”

“`in_memory`: Entries typically remain active for around 5 to 10 minutes of inactivity, up to one hour.”

“`24h`: Extended retention typically keeps entries available for around 30 minutes and can retain them for up to 24 hours.”

“reusing the prefix refreshes its lifetime without another cache-write charge.”

“The minimum cacheable prompt length is 1,024 tokens for GPT-5.6 and later and varies by request settings for earlier models.”

“For GPT-5.6 and later, cache writes cost 1.25× the standard, uncached input-token rate. Subsequent reads cost 0.1× that rate on most of these models and 0.05× on GPT-6.1 Sol.”

“| Cache write charge | 1.25× the uncached input-token rate | No additional cache-write charge | No additional cache-write charge |”

“Cache-write pricing is not an additive fee: input tokens use the uncached-input, cached-input, or cache-write rate.”

“Set `prompt_cache_options.mode` to `explicit` to use only developer-selected breakpoints and mark each desired breakpoint by adding `prompt_cache_breakpoint: { "mode": "explicit" }` to a supported content block inside an input message.”

“Each request can create up to four cache writes.”

“Better cost efficiency: 50% cost discount compared to synchronous APIs”

“Each batch completes within 24 hours (and often more quickly)”

“For now, the completion window can only be set to `24h`.”

“A single batch may include up to 50,000 requests, and a batch input file can be up to 200 MB in size.”

“You can create up to 2,000 batches per hour.”

“You will be charged for tokens consumed from any completed requests.”