> GPT-5.6 Luna is a closed AI model from OpenAI, released 9 Jul 2026. Index score 169 (GPT-4 = 100), rank 23 of 229 ranked models. API price $0.20 in / $1.2 out per million tokens, 1.05M context.

[Models](https://themodelindex.org/models/) /[OpenAI](https://themodelindex.org/labs/openai/)

# GPT-5.6 Luna

OpenAI

Released 9 Jul 2026

Closed

Flagship

[Compare](https://themodelindex.org/compare/gpt-5-6-luna-vs-kimi-k3/) [Announcement](https://openai.com/index/improving-gpt-5-6-sol-in-chatgpt/)

169158–180

Index score

#23 of 229 models

3/ 8

Core tests taken

3 of 6 areas · plus 16 anchor tests

−16

Versus the frontier at release

best then: Claude Fable 5

Replaced by [GPT-6 Luna](https://themodelindex.org/models/gpt-6-luna/)

$0.20/ $1.2

Price per million tokens

input / output, list price

1.05M

Context window

tokens it reads at once

## Where it sits · on the frontier

_Chart: Model Index score of every scored model since late 2022, with the best closed and best open model over time_

GPT-5.3-Codex reached this score 5 months earlier.

## Test results · by area · tick = best by any model

Show 14 anchor tests

Reasoning & knowledge

evidence

GPQA Diamond

92%

Humanity's Last Exam

no result

Maths

evidence

FrontierMath T1-3

82%

Coding

evidence

SWE-bench Verified

no result

Terminal-Bench

no result

Agents

evidence

APEX-Agents

no result

Novel problems

evidence

ARC-AGI-2

60%

Long tasks

no evidence yet

METR time horizon

no result

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

## What others say · rank on each leaderboard

The Model Index

independent tests, our method

#23

of 219 · 169

Epoch Capabilities Index

overall capability from many benchmarks

#21

of 239 · 156.4

LMArena · Maths

maths questions in chat

#40

of 373 · 1475

LMArena · Text

overall chat quality, judged by people

#65

of 388 · 1453

LMArena · Coding

coding questions in chat

#67

of 383 · 1502

LMArena · Creative writing

creative writing

#83

of 386 · 1409

LMArena · Vision

understanding images

#38

of 149 · 1259

LMArena · WebDev

building web apps and front ends

#44

of 123 · 1518

LMArena · Documents

working with long documents

#21

of 41 · 1457

LiveBench · Overall

overall, with fresh questions to limit contamination

#39

of 58 · 73.6

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.

## Best at · top three among current models

Replaced by GPT-6 Luna; it no longer leads any category among current models.

Where this model places in the top three of today's models, on category leaderboards and on our core tests.

## Where to use it · prices checked 1 Oct 2026

#### From the lab

ChatGPT

app · subscription

↗

OpenAI API

API

↗

#### API providers · 3 · per million tokens, in / out

O

OpenAI

$0.40 / $2.4

cheapest

A

Azure

$0.22 / $1.32

↗

A

Amazon Bedrock

$0.22 / $1.32

↗

Provider prices come from OpenRouter's live listing; going direct to a provider can differ. Subscriptions are the lab's own apps.

## Facts and sources

Released

9 Jul 2026

Announcement

[openai.com ↗](https://openai.com/index/improving-gpt-5-6-sol-in-chatgpt/)

Release entry

GPT-5.6 (Sol/Terra/Luna; preview 06-26)

Scored as

GPT-5.6 Luna

Artificial Analysis index

–

Output speed

–

Hugging Face

closed weights

## OpenAI releases · around this one

GPT-5.4 Thinking

Closed

5 MAR 2026

GPT-5.5 Thinking

Closed

23 APR 2026

GPT-5.6 Luna

Closed

9 JUL 2026

index 169

GPT-5.6 Sol

Closed

9 JUL 2026

index 184

GPT-5.6 Terra

Closed

9 JUL 2026

index 178

---
Source: https://themodelindex.org/models/gpt-5-6-luna/ · The Model Index · data as of 2026-10-01