GPT-5 nano
$0.050 per million input tokens, $0.40 per million output.
| Input | $0.050 / MTok |
|---|---|
| Output | $0.40 / MTok |
| Cache read | $0.005 / MTok |
| Cache write (5 min) | billed at the base input rate |
| Context window | not published here |
| Tokenizer family | o200k_base (OpenAI) |
What that costs per month
Monthly spend for a prompt of a given size, with a 500-token response, no caching. Rows are prompt size; columns are requests per day.
| Prompt size | 1,000 / day | 10,000 / day | 100,000 / day |
|---|---|---|---|
| 1,000 tokens | $7.60 | $76.04 | $760.42 |
| 5,000 tokens | $13.69 | $136.88 | $1,369 |
| 20,000 tokens | $36.50 | $365.00 | $3,650 |
Prompt caching on GPT-5 nano
A 10,000-token static prefix at 10,000 requests a day costs $220.52 a month uncached. With caching at an 80% hit rate it costs $111.02 — a saving of $109.50, or 50%. Caching pays for itself after 0 reads of the same prefix.
How cache economics workPrice your actual prompt
These figures assume a prompt size. Paste your real one and the analyser will count it with GPT-5 nano’s tokenizer family, find what is wasted, and compare against every other model at your volume.
Open the analyserPrices last verified 2026-08-10. Confirm against OpenAI’s own pricing page before committing to anything.