All times Eastern (ET)Provider pages checked daily at
ProSign inRSS

GPT-5.6 Luna

API ID gpt-5.6-luna

gpt-5.6-luna is an OpenAI model, released July 2026. It has a 1,050,000-token context window and up to 128,000 output tokens. It takes image input and supports function calling, structured outputs and reasoning. Standard price: from $0.20 input and from $1.20 output per 1M tokens.

Status
Active
Released
July 2026
Shutdown
none announced
Replacement
—

No deprecation or shutdown date has been announced as of October 15, 2026.

Free tier: Free for shared data, 10M tokens a day for Launch and Grow (2.5M for Build), in place when tracking began (October 12, 2026) (source) · All free models

Benchmarks

Scores as each source published them
or

Comparing with GPT-5 nano (2025-08-07), the previous model in this line. A score is compared only with the same benchmark from the same source, at the same setting where both have it. Full comparison: prices, limits, features

Arena

ArenaGPT-5.6 LunaGPT-5 nano (2025-08-07)ChangeRankDateSource
Text
By category
  • Math 1478 rank 42 vs not rated
  • Creative writing 1410 rank 91 vs not rated
  • Instruction following 1443 rank 76 vs not rated
  • Hard prompts 1474 rank 79 vs not rated
  • Long prompts 1452 rank 89 vs not rated
  • Multi-turn 1457 rank 79 vs not rated
  • Expert 1490 rank 60 vs not rated
1453 ±4not rated75 of 415
41,103 votes		Oct 13, 2026source
Text: Coding1501 ±7not rated77 of 410
11,126 votes		Oct 13, 2026source
WebDev1518 ±6not rated52 of 142
12,656 votes		Oct 13, 2026source
Vision1258 ±7not rated48 of 165
11,086 votes		Oct 9, 2026source
Document1462 ±8not rated23 of 44
5,766 votes		Sep 13, 2026source
Agent-0.018 ±0.007not rated37 of 54
43,217 sessions		Oct 14, 2026source

Rated as gpt-5.6-luna-xhigh (xhigh effort). Text, WebDev, Vision and Document ratings come from people voting between two anonymous answers; Text, Vision and Document are style-controlled, as on arena.ai. Agent is Arena's separate score for real agent sessions, on its own scale. Data: Arena (arena.ai), CC BY 4.0.

Epoch AI

BenchmarkGPT-5.6 LunaGPT-5 nano (2025-08-07)ChangeSettingRunSource
FrontierMath Tier 461.0%2.4%+58.6 ptsmax effort (vs high effort)Jul 9, 2026source
FrontierMath Tiers 1-3
Other settings
  • 41.4% vs 6.0% low effort
  • 39.6% vs not run no reasoning
82.1%20.0%+62.1 ptsmax effort (vs high effort)Jul 9, 2026source
GPQA Diamond
Other settings
  • 82.3% vs 57.6% low effort
  • 63.6% vs not run no reasoning
91.6%69.4%+22.2 ptsmax effort (vs high effort)Jul 9, 2026source
OTIS Mock AIME 2024-2025
Other settings
  • 66.7% vs 46.7% low effort
  • 40.0% vs not run no reasoning
98.3%81.1%+17.2 ptsmax effort (vs high effort)Jul 9, 2026source
SimpleQA Verified41.0%11.7%+29.3 ptsmax effort (vs high effort)Aug 10, 2026source

Run independently by Epoch AI. Data: Epoch AI, Capabilities & benchmarking (epoch.ai/benchmarks), CC BY.

On cloud platforms

All platform dates
PlatformPlatform IDStateShutdownTime leftReplacementSource
Amazon Bedrockopenai.gpt-5.6-lunaActivenot before Jul 13, 2027—source
Microsoft Foundryversion 2026-07-09gpt-5.6-lunaActiveJan 11, 2028453 days—source

Read daily from each platform's own pages. The shutdown date and countdown on this page use OpenAI's own date. Time left counts to 00:00 UTC on the shutdown date, which is the evening before in ET.

Features

Compare models
FeatureValueSinceSource
Structured outputsThe model can be held to a JSON schema you supply, so every reply parses without retries or cleanup.Supportedin place when tracking began (October 7, 2026)source
Function callingAlso called tool use. You describe functions; the model replies with a function name and arguments for your code to run, then uses the result.Supportedin place when tracking began (October 7, 2026)source
Prompt cachingA prompt prefix you send again and again (system prompt, documents, tools) is cached, so later requests cost less and start faster.Supportedin place when tracking began (October 7, 2026)source
Batch APISend many requests as one job that runs in the background, usually finished within 24 hours, at a lower price than live requests.Supportedin place when tracking began (October 7, 2026)source
Image inputImages such as screenshots, photos and charts can go in the prompt for the model to read.Supportedin place when tracking began (October 7, 2026)source
Reasoning / thinkingThe model can work through a problem step by step before it answers: usually better answers to hard questions, at the cost of more tokens and time.Supportedin place when tracking began (October 7, 2026)source
Fine-tuningYou can train your own version of the model on your examples through the company's API, then call it like any other model.Not supportedin place when tracking began (October 7, 2026)source
Built-in web searchThe company's API can search the web for the model during a request and cite its sources; you don't need a search service of your own.Supportedin place when tracking began (October 7, 2026)source
Context windowThe most the model can work with in one request, in tokens: your prompt, conversation history and files. A token is about three-quarters of an English word.1,050,000 tokensin place when tracking began (October 7, 2026)source
Max output tokensThe longest reply the model can write in one response, in tokens. A token is about three-quarters of an English word.128,000 tokensin place when tracking began (October 7, 2026)source

Not published by OpenAI: PDF input, temperature parameter, top_p parameter, logprobs.

Price history

USD per 1M tokens, standard tier
$0.00$0.50$1.00$1.50Oct 2026In $0.20Out $1.20Oct 1, 2026 to Oct 15, 2026: Input $0.20, Output $1.20
$0.00$0.50$1.00$1.50Oct 2026In $0.20Out $1.20Oct 1, 2026 to Oct 15, 2026: Input $0.20, Output $1.20

We started recording this model's prices on Oct 1, 2026. The history grows with each daily check.

Show as a table
FromTypePrice
Oct 1, 2026Input$0.20
Oct 1, 2026Output$1.20

Prices

USD per 1M tokens, every type and band, by tier
TypeBandPrice $/1MIn effect
Standard
inputlong context$0.40
inputshort context$0.20
cached inputlong context$0.04
cached inputshort context$0.02
cache writelong context$0.50
cache writeshort context$0.25
outputlong context$1.80
outputshort context$1.20
Batch
inputlong context$0.20
inputshort context$0.10
cached inputlong context$0.02
cached inputshort context$0.01
cache writelong context$0.25
cache writeshort context$0.125
outputlong context$0.90
outputshort context$0.60
Flex
inputlong context$0.20
inputshort context$0.10
cached inputlong context$0.02
cached inputshort context$0.01
cache writelong context$0.25
cache writeshort context$0.125
outputlong context$0.90
outputshort context$0.60
Fast
inputlong context$0.80
inputshort context$0.40
cached inputlong context$0.08
cached inputshort context$0.04
cache writelong context$1.00
cache writeshort context$0.50
outputlong context$3.60
outputshort context$2.40

Changelog

Changes we detected on OpenAI's pages

No changes recorded yet.

Email alerts

Get an email if gpt-5.6-luna changes

We'll email you the same morning we spot a deprecation notice, a new or moved shutdown date, a price change, or the model leaving OpenAI's model list.

No password: one email to confirm, and you can unsubscribe any time. Privacy

Last checked · Sources: OpenAI deprecations, OpenAI pricing, OpenAI models