# GPT-5.6 Luna API pricing

GPT-5.6 Luna is an earlier OpenAI model, still sold. It costs $0.20 per million tokens it reads and $1.20 per million it writes, $0.45 on a mix of 3 read for each 1 written. OpenAI's newest models are GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna. Below are every price OpenAI lists for it, and what 8 everyday jobs cost a month on it.

- Page: https://tecyfy.com/prices/gpt-5-6-luna
- Maker: OpenAI

## Price per million tokens, in US dollars

| | Price |
| --- | --: |
| Input | $0.20 |
| Cached input | $0.02 |
| Cache write | $0.25 |
| Output | $1.20 |
| Blend, 3 read for each 1 written | $0.45 |
| Batch input | $0.10 |
| Batch output | $0.60 |
| Long prompt input (over 272K tokens) | $0.40 |
| Long prompt output | $1.80 |
| Context window, tokens | 1.05M |
| Longest answer, tokens | 128K |
| Capability score (Epoch Capabilities Index; GPT-5 is 150) | 156 |

Price change: The price fell by 80%.

## What 8 everyday jobs cost a month

Each job at its usual runs a day, for 30 days, at the standard price.

| Job | Tokens in, out, runs a day | A run | A month |
| --- | --- | --: | --: |
| A chat question | 300, 500, 100 | 0.066 cents | $1.98 |
| A hard question, with thinking | 300, 1,400, 100 | 0.17 cents | $5.22 |
| A long report, summarised | 60,000, 2,000, 20 | 1.4 cents | $8.64 |
| A 40-turn agent task | 2,800,000, 40,000, 10 | 15 cents | $46.32 |
| A support bot's reply | 4,000, 300, 2,000 | 0.08 cents | $48.00 |
| Sort a message into a category | 250, 5, 50,000 | 0.0056 cents | $84.00 |
| Read an invoice into fields | 1,500, 250, 1,000 | 0.06 cents | $18.00 |
| Review a pull request | 25,000, 4,500, 50 | 1 cent | $15.60 |

## Cheaper models that score as well as GPT-5.6 Luna

- [Claude Haiku 5.5](https://tecyfy.com/prices/claude-haiku-5-5.md), Anthropic: $0.20 per million at 3:1, score 159
- [GPT-6 Luna](https://tecyfy.com/prices/gpt-6-luna.md), OpenAI: $0.20 per million at 3:1, score 156

## Questions

### How much does GPT-5.6 Luna cost?

GPT-5.6 Luna costs $0.20 per million input tokens, the text it reads, and $1.20 per million output tokens, the text it writes, thinking included. On a mix of 3 tokens read for each 1 written, that is $0.45 per million.

### How much does cached input cost on GPT-5.6 Luna?

Text GPT-5.6 Luna reads again from the cache costs $0.02 per million tokens, against $0.20 the first time. Putting text in the cache costs $0.25 per million, so a long prompt sent again and again costs far less after the first time.

### Does GPT-5.6 Luna have a batch price?

Yes. With a batch, where you send many requests and collect the answers later instead of at once, GPT-5.6 Luna costs $0.10 per million tokens read and $0.60 per million written.

### Does a long prompt cost more on GPT-5.6 Luna?

Yes. Once a prompt passes 272K tokens, the whole request costs $0.40 per million tokens read and $1.80 per million written, against $0.20 and $1.20 below that.

### What is the context window of GPT-5.6 Luna?

GPT-5.6 Luna holds 1.05M tokens at once, about 1,600 pages of English text, and writes up to 128K tokens in one answer. The window holds your instructions, the files you attach, the conversation so far and the answer.

### How good is GPT-5.6 Luna?

GPT-5.6 Luna scores 156 on the Epoch Capabilities Index, an independent score from Epoch AI that combines over 50 benchmarks, with a range of doubt from 154 to 159 (GPT-5 is 150). One score is an average; test it on 10 tasks of your own.

### Is there a cheaper model as good as GPT-5.6 Luna?

Claude Haiku 5.5 from Anthropic scores 159 on the Epoch Capabilities Index, about as well or better, and costs $0.20 per million on a mix of 3 tokens read for each 1 written: GPT-5.6 Luna costs 2.3 times as much. Test both on your own work before you switch.

### Should I still use GPT-5.6 Luna?

It is still sold, so it keeps working, but OpenAI's newest models are GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna. Compare its price with theirs below, and run your own tasks on the newer model before you move.

### What does a support bot cost a month on GPT-5.6 Luna?

At 2,000 replies a day, each reading 4,000 tokens (50 percent of them from cache) and writing 300, a support bot costs 0.08 cents a reply on GPT-5.6 Luna, $48.00 a month.

### What does an AI agent task cost on GPT-5.6 Luna?

One 40-turn task that reads 2,800,000 tokens in all, 90 percent of them from cache, and writes 40,000, costs 15 cents on GPT-5.6 Luna. At 10 tasks a day that is $46.32 a month.

## Sources

- [OpenAI: API prices](https://developers.openai.com/api/docs/pricing)
- [Epoch Capabilities Index, by Epoch AI (CC BY 4.0)](https://epoch.ai/benchmarks/eci)
- Every model's prices as [JSON](https://tecyfy.com/prices.json) or [CSV](https://tecyfy.com/prices.csv), free to use

## Compare

- [GPT-5.6 Luna vs GPT-6 Astra](https://tecyfy.com/prices/compare/gpt-5-6-luna-vs-gpt-6-astra.md)
- [GPT-5.6 Luna vs GPT-6.1 Sol](https://tecyfy.com/prices/compare/gpt-5-6-luna-vs-gpt-6-1-sol.md)
- [GPT-5.6 Luna vs GPT-6 Luna](https://tecyfy.com/prices/compare/gpt-5-6-luna-vs-gpt-6-luna.md)

## OpenAI's other models

- [GPT-6 Astra](https://tecyfy.com/prices/gpt-6-astra.md)
- [GPT-6.1 Sol](https://tecyfy.com/prices/gpt-6-1-sol.md)
- [GPT-6 Luna](https://tecyfy.com/prices/gpt-6-luna.md)
- [GPT-6 Sol](https://tecyfy.com/prices/gpt-6-sol.md)
- [GPT-5.6 Sol](https://tecyfy.com/prices/gpt-5-6-sol.md)
- [GPT-5.6 Terra](https://tecyfy.com/prices/gpt-5-6-terra.md)
- [GPT-5.5](https://tecyfy.com/prices/gpt-5-5.md)
- [GPT-5.5 pro](https://tecyfy.com/prices/gpt-5-5-pro.md)
- [GPT-5.4](https://tecyfy.com/prices/gpt-5-4.md)
- [GPT-5.4 mini](https://tecyfy.com/prices/gpt-5-4-mini.md)
- [GPT-5.4 nano](https://tecyfy.com/prices/gpt-5-4-nano.md)
- [GPT-5.4 pro](https://tecyfy.com/prices/gpt-5-4-pro.md)
- [GPT-5.2](https://tecyfy.com/prices/gpt-5-2.md)
- [GPT-5.2 pro](https://tecyfy.com/prices/gpt-5-2-pro.md)
- [GPT-5.1](https://tecyfy.com/prices/gpt-5-1.md)
- [GPT-5](https://tecyfy.com/prices/gpt-5.md)
- [GPT-5 mini](https://tecyfy.com/prices/gpt-5-mini.md)
- [GPT-5 nano](https://tecyfy.com/prices/gpt-5-nano.md)
- [GPT-5 pro](https://tecyfy.com/prices/gpt-5-pro.md)
- [GPT-4.1](https://tecyfy.com/prices/gpt-4-1.md)
- [GPT-4.1 mini](https://tecyfy.com/prices/gpt-4-1-mini.md)
- [GPT-4.1 nano](https://tecyfy.com/prices/gpt-4-1-nano.md)
- [GPT-4o](https://tecyfy.com/prices/gpt-4o.md)
- [GPT-4o (2024-05-13)](https://tecyfy.com/prices/gpt-4o-2024-05-13.md)
- [GPT-4o mini](https://tecyfy.com/prices/gpt-4o-mini.md)
- [o1](https://tecyfy.com/prices/o1.md)
- [o1-pro](https://tecyfy.com/prices/o1-pro.md)
- [o3-pro](https://tecyfy.com/prices/o3-pro.md)
- [o3](https://tecyfy.com/prices/o3.md)
- [o4-mini](https://tecyfy.com/prices/o4-mini.md)
- [o3-mini](https://tecyfy.com/prices/o3-mini.md)
- [GPT-4 Turbo (2024-04-09)](https://tecyfy.com/prices/gpt-4-turbo-2024-04-09.md)
- [GPT-4 (0613)](https://tecyfy.com/prices/gpt-4-0613.md)
- [GPT-3.5 Turbo](https://tecyfy.com/prices/gpt-3-5-turbo.md)
- [GPT-3.5 Turbo (0125)](https://tecyfy.com/prices/gpt-3-5-turbo-0125.md)
- [GPT-3.5 Turbo (1106)](https://tecyfy.com/prices/gpt-3-5-turbo-1106.md)
- [GPT-3.5 Turbo Instruct](https://tecyfy.com/prices/gpt-3-5-turbo-instruct.md)
- [davinci-002](https://tecyfy.com/prices/davinci-002.md)
- [babbage-002](https://tecyfy.com/prices/babbage-002.md)