Model pricing
GPT-5.6 Luna API list price — cost per token
GPT-5.6 Luna costs $0.20 per million input tokens and $1.20 per million output tokens (USD). Requests with >272K tokens in the prompt, including cache reads and writes, use the premium rates for the whole request.
| Token rate | USD / 1M tokens |
|---|---|
| Effortqualitative reasoning depth | High |
| Inputper 1M tokens | $0.20 |
| Output | $1.20 |
| Cache writeper 1M tokens | $0.25 |
| Cache readper 1M tokens | $0.02 |
| Input · >272K context | $0.40 |
| Output · >272K context | $1.80 |
| Cache write · >272K context | $0.50 |
| Cache read · >272K context | $0.04 |
Notional list price. These are OpenAI (GPT / Codex) API list prices in USD. On Codex usage this is a notional list price — what usage would cost at API rates, not metered spend.
Cache & context economics
Cache reads cost $0.02 per million tokens — about a tenth of input. Cache writes cost $0.25 per million tokens. The >272K tokens context tier reprices the whole request.
Common questions
- How much does GPT-5.6 Luna cost per million tokens?
- $0.20 input and $1.20 output per million tokens (USD list price).
- What is the cache-read price for GPT-5.6 Luna?
- $0.02 per million cached-read tokens — about a tenth of the input rate.
- How is cache writing priced for GPT-5.6 Luna?
- Cache writes cost $0.25 per million tokens.
- Does GPT-5.6 Luna have a context pricing tier?
- Yes — at >272K tokens input rises to $0.40 per million for the entire request.
Track your real cost with JP the Cat
A live menu-bar total, priced to the cent from your local logs — including the cache-write split most tools miss.
Download MeowPrefer Homebrew? Install the same signed, notarized app from JP's public tap.
brew tap 8bittts/jpthecat
brew trust 8bittts/jpthecat
brew install --cask jpthecat