Where is the GPT-5.6 Luna documentation?

The GPT-5.6 Luna documentation on OpenRouter includes the GPT-5.6 Migration Guide cookbook and the GPT 5.6 Discounts & Jevons Paradox blog post. The quickstart covers API keys and your first request, and the API reference lists every parameter.

How much does GPT-5.6 Luna cost?

GPT-5.6 Luna costs $0.20/M input tokens and $1.20/M output tokens, with separate rates for Cache Read at $0.02/M tokens and Cache Write at $0.25/M tokens.

What is the context length of GPT-5.6 Luna?

GPT-5.6 Luna has a 1,050,000 token context window. It supports up to 128,000 completion tokens.

Does GPT-5.6 Luna support tool calling and structured outputs?

Yes. GPT-5.6 Luna accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

Can I disable prompt caching for GPT-5.6 Luna?

Yes. OpenAI caches prompts of 1024 tokens or more automatically, and GPT-5.6 Luna bills a cache write on each cached prompt whether or not the cached prefix is read again. To turn caching off for a request, set prompt_cache_options.mode to "explicit" at the request root and do not mark any content block with prompt_cache_breakpoint. Explicit mode disables OpenAI-managed breakpoints, so a request with none of your own has no cacheable prefix and is billed at the plain input rate. See the prompt caching guide.

What inputs and outputs does GPT-5.6 Luna support?

GPT-5.6 Luna accepts files such as PDFs, images, and text as input and returns text.

Which providers serve GPT-5.6 Luna?

GPT-5.6 Luna is served by 3 providers on OpenRouter: OpenAI, Amazon Bedrock, and Azure. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.

When was GPT-5.6 Luna released?

GPT-5.6 Luna was released on July 9, 2026. Its knowledge cutoff is February 16, 2026.

OpenAI: GPT-5.6 Luna

openai/gpt-5.6-luna

Compare

GPT-5.6 Luna is a fast, cost-efficient model in OpenAI's GPT-5.6 series. It is suited for high-volume, latency-sensitive tasks such as chat, classification, and lightweight agentic workflows, providing capable reasoning for its price tier.

Modalities

In / out price

$0.20 / $1.20per 1M

Context

1.1M

Released

Jul 9, 2026

Knowledge cutoff

Feb 2026

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).

Latency / throughput


	$0.20	$1.20	$0.02	1.37s	114 tok/s	100.00%
	$0.20	$1.20	$0.02	1.67s	62 tok/s	100.00%
	$0.20	$1.20	$0.02	3.48s	72 tok/s	97.70%
Not used in Standard routing:Why these endpoints are not used
Flex	$0.10	$0.60	$0.01	2.20s	82 tok/s	99.97%
	$0.22	$1.32	$0.022	1.23s	83 tok/s	100.00%
	$0.22	$1.32	$0.022	0.37s	158 tok/s	98.45%
	$0.22	$1.32	$0.022	1.36s	107 tok/s	100.00%
Fast	$0.40	$2.40	$0.04	0.86s	89 tok/s	100.00%

Input $0.20Output $1.20
114 tok/s100.00% uptime
Input $0.20Output $1.20
62 tok/s100.00% uptime
Input $0.20Output $1.20
72 tok/s97.70% uptime
Not used in Standard routing:Why these endpoints are not used
Input $0.10Output $0.60
Flex
82 tok/s99.97% uptime
Input $0.22Output $1.32
83 tok/s100.00% uptime
Input $0.22Output $1.32
158 tok/s98.45% uptime
Input $0.22Output $1.32
107 tok/s100.00% uptime
Input $0.40Output $2.40
Fast
89 tok/s100.00% uptime

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Weighted average

Input

$0.1232/M tokens

Output

$1.187/M tokens

GPT-5.6 Luna — Price history

Price history

No effective price data for this selection

		$0.0470	$1.201	89.6%	18.6%
		$0.1386	$1.199	48.4%	75.8%
		$0.2119	$1.199	16.5%	2.4%
Not used in Standard routing:Why these endpoints are not used
	Flex	$0.0533	$0.599	58.7%	1.2%
		$0.1373	$1.319	54.2%	0.1%
		$0.1451	$1.319	50.7%	1.5%
	Fast	$0.1767	$2.399	69.8%	0.1%
		$0.1783	$1.318	37.5%	0.3%

Effective in $0.0470Effective out $1.201
Token share
18.6%
Effective in $0.1386Effective out $1.199
Token share
75.8%
Effective in $0.2119Effective out $1.199
Token share
2.4%
Not used in Standard routing:Why these endpoints are not used
Effective in $0.0533Effective out $0.599
FlexToken share
1.2%
Effective in $0.1373Effective out $1.319
Token share
0.1%
Effective in $0.1451Effective out $1.319
Token share
1.5%
Effective in $0.1767Effective out $2.399
FastToken share
0.1%
Effective in $0.1783Effective out $1.318
Token share
0.3%

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Uptime (3d)

100.00%

Availability (3d)

98.99%

Availability over the last 3 days

Last 72 hours

Availability 98.99%

3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability

96.79%

Without Routing

94.15%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for OpenAI: GPT-5.6 Luna (Artificial Analysis and Vals AI)
Source	Benchmark	Score
Artificial Analysis	GPT-5.6 Luna (Max) Intelligence Index	23.8
Artificial Analysis	GPT-5.6 Luna (Max) Coding Index	71.4
Artificial Analysis	GPT-5.6 Luna (Max) Agentic Index	42.1
Artificial Analysis	GPT-5.6 Luna (Max) GPQA Diamond	91.1%
Artificial Analysis	GPT-5.6 Luna (Max) HLE	39.5%
Artificial Analysis	GPT-5.6 Luna (Max) AA-LCR	83.7%
Artificial Analysis	GPT-5.6 Luna (Max) τ-Bench Banking	31.1%
Artificial Analysis	GPT-5.6 Luna (Max) GDPval-AA	48.4%
Artificial Analysis	GPT-5.6 Luna (Max) CritPt	20.6%
Artificial Analysis	GPT-5.6 Luna (Max) SciCode	53.6%
Artificial Analysis	GPT-5.6 Luna (Max) Terminal-Bench 4.0	11.6%
Artificial Analysis	GPT-5.6 Luna (Max) Terminal-Bench 2.1	80.9%
Artificial Analysis	GPT-5.6 Luna (Max) AA-Omniscience Accuracy	39.1%
Artificial Analysis	GPT-5.6 Luna (Max) AA-Omniscience Non-Hallucination Rate	7.7%
Artificial Analysis	GPT-5.6 Luna (Low) Coding Index	44.2
Artificial Analysis	GPT-5.6 Luna (Low) Agentic Index	16.1
Artificial Analysis	GPT-5.6 Luna (Low) GPQA Diamond	83.5%
Artificial Analysis	GPT-5.6 Luna (Low) HLE	19.8%
Artificial Analysis	GPT-5.6 Luna (Low) AA-LCR	70.0%
Artificial Analysis	GPT-5.6 Luna (Low) τ-Bench Banking	12.8%
Artificial Analysis	GPT-5.6 Luna (Low) GDPval-AA	25.1%
Artificial Analysis	GPT-5.6 Luna (Low) CritPt	2.6%
Artificial Analysis	GPT-5.6 Luna (Low) SciCode	46.1%
Artificial Analysis	GPT-5.6 Luna (Low) Terminal-Bench 4.0	0.0%
Artificial Analysis	GPT-5.6 Luna (Low) Terminal-Bench 2.1	43.4%
Artificial Analysis	GPT-5.6 Luna (Xhigh) Coding Index	68.6
Artificial Analysis	GPT-5.6 Luna (Xhigh) Agentic Index	38.7
Artificial Analysis	GPT-5.6 Luna (Xhigh) GPQA Diamond	89.5%
Artificial Analysis	GPT-5.6 Luna (Xhigh) HLE	37.0%
Artificial Analysis	GPT-5.6 Luna (Xhigh) AA-LCR	81.7%
Artificial Analysis	GPT-5.6 Luna (Xhigh) τ-Bench Banking	28.7%
Artificial Analysis	GPT-5.6 Luna (Xhigh) GDPval-AA	45.4%
Artificial Analysis	GPT-5.6 Luna (Xhigh) CritPt	20.6%
Artificial Analysis	GPT-5.6 Luna (Xhigh) SciCode	50.5%
Artificial Analysis	GPT-5.6 Luna (Xhigh) Terminal-Bench 4.0	3.5%
Artificial Analysis	GPT-5.6 Luna (Xhigh) Terminal-Bench 2.1	77.9%
Artificial Analysis	GPT-5.6 Luna (Medium) Coding Index	50.7
Artificial Analysis	GPT-5.6 Luna (Medium) Agentic Index	23.9
Artificial Analysis	GPT-5.6 Luna (Medium) GPQA Diamond	85.9%
Artificial Analysis	GPT-5.6 Luna (Medium) HLE	25.8%
Artificial Analysis	GPT-5.6 Luna (Medium) AA-LCR	75.0%
Artificial Analysis	GPT-5.6 Luna (Medium) τ-Bench Banking	17.7%
Artificial Analysis	GPT-5.6 Luna (Medium) GDPval-AA	31.9%
Artificial Analysis	GPT-5.6 Luna (Medium) CritPt	4.9%
Artificial Analysis	GPT-5.6 Luna (Medium) SciCode	46.8%
Artificial Analysis	GPT-5.6 Luna (Medium) Terminal-Bench 4.0	0.5%
Artificial Analysis	GPT-5.6 Luna (Medium) Terminal-Bench 2.1	53.2%
Artificial Analysis	GPT-5.6 Luna (Non-reasoning) Coding Index	39.3
Artificial Analysis	GPT-5.6 Luna (Non-reasoning) Agentic Index	13.4
Artificial Analysis	GPT-5.6 Luna (Non-reasoning) GPQA Diamond	64.5%
Artificial Analysis	GPT-5.6 Luna (Non-reasoning) HLE	7.2%
Artificial Analysis	GPT-5.6 Luna (Non-reasoning) AA-LCR	42.7%
Artificial Analysis	GPT-5.6 Luna (Non-reasoning) τ-Bench Banking	9.7%
Artificial Analysis	GPT-5.6 Luna (Non-reasoning) GDPval-AA	21.4%
Artificial Analysis	GPT-5.6 Luna (Non-reasoning) CritPt	0.3%
Artificial Analysis	GPT-5.6 Luna (Non-reasoning) SciCode	40.4%
Artificial Analysis	GPT-5.6 Luna (Non-reasoning) Terminal-Bench 4.0	1.0%
Artificial Analysis	GPT-5.6 Luna (Non-reasoning) Terminal-Bench 2.1	39.0%
Artificial Analysis	GPT-5.6 Luna (High) Coding Index	63.3
Artificial Analysis	GPT-5.6 Luna (High) Agentic Index	34.6
Artificial Analysis	GPT-5.6 Luna (High) GPQA Diamond	89.2%
Artificial Analysis	GPT-5.6 Luna (High) HLE	33.4%
Artificial Analysis	GPT-5.6 Luna (High) AA-LCR	80.3%
Artificial Analysis	GPT-5.6 Luna (High) τ-Bench Banking	25.2%
Artificial Analysis	GPT-5.6 Luna (High) GDPval-AA	41.9%
Artificial Analysis	GPT-5.6 Luna (High) CritPt	16.6%
Artificial Analysis	GPT-5.6 Luna (High) SciCode	51.6%
Artificial Analysis	GPT-5.6 Luna (High) Terminal-Bench 4.0	2.5%
Artificial Analysis	GPT-5.6 Luna (High) Terminal-Bench 2.1	69.7%
Vals AI	GPT-5.6 Luna Access to Justice Benchmark	65.6%
Vals AI	GPT-5.6 Luna BioMysteryBench	61.5%
Vals AI	GPT-5.6 Luna Code Migration	44.5%
Vals AI	GPT-5.6 Luna CorpFin v2	64.2%
Vals AI	GPT-5.6 Luna CyberBench v1.1	73.6%
Vals AI	GPT-5.6 Luna EMB	67.1%
Vals AI	GPT-5.6 Luna Finance Agent (v2)	55.0%
Vals AI	GPT-5.6 Luna Financial Coaching Bench	69.3%
Vals AI	GPT-5.6 Luna GPQA Diamond	91.7%
Vals AI	GPT-5.6 Luna Harvey's Legal Agent Benchmark v1	1.3%
Vals AI	GPT-5.6 Luna IOI	61.8%
Vals AI	GPT-5.6 Luna LegalBench	84.0%
Vals AI	GPT-5.6 Luna Legal Research Bench	36.5%
Vals AI	GPT-5.6 Luna MedCode	42.4%
Vals AI	GPT-5.6 Luna MedScribe	84.4%
Vals AI	GPT-5.6 Luna MMLU Pro	86.0%
Vals AI	GPT-5.6 Luna MMMU Pro	85.0%
Vals AI	GPT-5.6 Luna MortgageTax	67.3%
Vals AI	GPT-5.6 Luna MysteryMechanism	14.4%
Vals AI	GPT-5.6 Luna ProgramBench	0.0%
Vals AI	GPT-5.6 Luna ProofBench v1.1	60.0%
Vals AI	GPT-5.6 Luna Public Benefits Bench v1.1	61.2%
Vals AI	GPT-5.6 Luna SAGE v1	44.2%
Vals AI	GPT-5.6 Luna Vals Science Index	25.3%
Vals AI	GPT-5.6 Luna SkillsBench	60.4%
Vals AI	GPT-5.6 Luna SRE Bench	1.5%
Vals AI	GPT-5.6 Luna SWE-bench	93.0%
Vals AI	GPT-5.6 Luna Tax Agent Bench	24.8%
Vals AI	GPT-5.6 Luna TaxEval v2	76.2%
Vals AI	GPT-5.6 Luna Terminal-Bench 4.0	11.6%
Vals AI	GPT-5.6 Luna Terminal-Bench-Science	0.0%
Vals AI	GPT-5.6 Luna Vals Index	51.7%
Vals AI	GPT-5.6 Luna Vals Multimodal Index	69.1%
Vals AI	GPT-5.6 Luna Vibe Code Bench 1-100	22.6%
Vals AI	GPT-5.6 Luna Vibe Code Bench v1.1	77.1%

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

Codex

A coding agent that helps you build and ship with AI

415Btokens

Sentry MCP

new

167Btokens

Hermes Agent

Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.

142Btokens

Fan Auto

new

83.4Btokens

Ito

Agentic QA platform that actually runs your app

53.4Btokens

Rest of OpenRouter

61.8B

Top 5 Apps

5.44B

OpenAI: GPT Luna versions

Available through EU in-region routing

OpenAI: GPT-5.6 Luna is available through EU in-region routing. Requests sent to eu.openrouter.ai are decrypted in the EU and routed only to shared provider endpoints in Europe (the EU/EEA, the UK, and Switzerland).

Providers in Europe

Azure

Prices in Europe

Azure: $0.22/M tokens input, $1.32/M tokens output

Base URL

https://eu.openrouter.ai/api/v1

Plan

Requires Business or Enterprise. Self-serve on Business: no contract, no monthly minimum. OpenRouter Business for EU in-region routing

Fails closed: if no shared provider endpoint in Europe is available, the request errors instead of leaving the region. Requests served by your own BYOK key go to your deployment’s region, which you control. In-region routing docs

More models from OpenAI

GPT Image 2.6 Sunburst

GPT Image 2.6 Sunburst is OpenAI's image generation and editing model optimized for detail: dense compositions, text-heavy layouts, and edits that keep details consistent from one edit to the next at medium or high quality, generating about 50% faster than GPT Image 2.5. It takes text and image inputs and produces images, with quality tiers from low to max, transparent backgrounds, and custom sizes in multiples of 16 pixels at aspect ratios from 1:3 to 3:1. GPT Image 2.6 introduces a complexity setting for how much the model plans before rendering, markup-guided edits, animated GIF output, and SVG output. Choose Flare instead when speed matters more than fine detail.

Image$8 / $30

GPT Image 2.6 Flare

GPT Image 2.6 Flare is OpenAI's image generation and editing model optimized for speed: fast iteration and interactive text-to-image and image-to-image workflows, generating about 50% faster than GPT Image 2.5. It takes text and image inputs and produces images, with quality tiers from low to max, transparent backgrounds, and custom sizes in multiples of 16 pixels at aspect ratios from 1:3 to 3:1. GPT Image 2.6 introduces a complexity setting for how much the model plans before rendering, markup-guided edits, animated GIF output, and SVG output. Choose Sunburst instead when detail and edit-to-edit consistency matter more than latency.

Image$8 / $30

GPT-6.1 Terra Pro

GPT-6.1 Terra Pro is the same underlying model as GPT-6.1 Terra, served with reasoning.mode set to pro for higher-quality responses on complex tasks.

Cost note: pro mode spends far more reasoning tokens per request, so a typical request costs several times more than the same request on GPT-6.1 Terra and takes much longer to complete. It is intended for hard, high-stakes problems where the extra accuracy justifies the cost. For everyday coding, agentic, and chat workloads, use GPT-6.1 Terra instead.

Learn more in OpenAI's docs: https://developers.openai.com/api/docs/guides/reasoning#reasoning-mode

Text1.1M context$1 / $5

GPT-6.1 Terra Pro (batch)

GPT-6.1 Terra Pro is the same underlying model as GPT-6.1 Terra, served with reasoning.mode set to pro for higher-quality responses on complex tasks.

Learn more in OpenAI's docs: https://developers.openai.com/api/docs/guides/reasoning#reasoning-mode

Text1.1M context$0.50 / $2.50

GPT-6.1 Terra

GPT-6.1 Terra is the mid-tier model in OpenAI's GPT-6.1 series, priced below GPT-6.1 Sol. It is suited for everyday agentic, coding, and writing work, is particularly strong at understanding screenshots and on-screen content, and is efficient enough for high-volume workloads.

Text1.1M context$1 / $5

GPT-6.1 Terra (batch)

Text1.1M context$0.50 / $2.50

GPT-6 Luna Decisions

GPT-6 Luna Decisions is GPT-6 Luna served through OpenAI's Decisions API. Instead of generating text, it reads the content passed as state (text, JSON, or images) and returns typed, probabilistic answers to named questions in a single request: the probability of yes for a yes/no question (noul), a probability for every option plus the most likely one (choice), or a probability for every level of an ordered rubric plus the expected score (score).

It is built for classification, routing, moderation, and rubric grading where application code thresholds the returned numbers rather than parsing a chat reply. A request can carry up to 200 questions about the same content.

Decisions$0.10 / $0

GPT-6.1 Sol Pro

GPT-6.1 Sol Pro is the same underlying model as GPT-6.1 Sol, served with reasoning.mode set to pro for higher-quality responses on complex tasks.

Cost note: pro mode spends far more reasoning tokens per request, so a typical request costs several times more than the same request on GPT-6.1 Sol and takes much longer to complete. It is intended for hard, high-stakes problems where the extra accuracy justifies the cost. For everyday coding, agentic, and chat workloads, use GPT-6.1 Sol instead.

Learn more in OpenAI's docs: https://developers.openai.com/api/docs/guides/reasoning#reasoning-mode

Text1.1M context$2 / $10

GPT-6.1 Sol Pro (batch)

GPT-6.1 Sol Pro is the same underlying model as GPT-6.1 Sol, served with reasoning.mode set to pro for higher-quality responses on complex tasks.

Learn more in OpenAI's docs: https://developers.openai.com/api/docs/guides/reasoning#reasoning-mode

Text1.1M context$1 / $5

GPT-6.1 Sol

GPT-6.1 Sol is an upgrade to GPT-6 Sol from OpenAI, positioned below the flagship GPT-6 Astra in the GPT-6 series. It is suited for agentic coding, computer use, document-heavy professional work, and multi-step business workflow automation, and approaches Astra-level results on these tasks at a much lower cost. Compared with GPT-6 Sol, it makes fewer factual errors and is more reliable at respecting explicit restrictions and user intent during agentic tasks.

Text1.1M context$2 / $10

GPT-6.1 Sol (batch)

Text1.1M context$1 / $5

GPT-6 Luna Pro

GPT-6 Luna Pro is the same underlying model as GPT-6 Luna, served with reasoning.mode set to pro for higher-quality responses on complex tasks.

Cost note: pro mode spends far more reasoning tokens per request, so a typical request costs several times more than the same request on GPT-6 Luna and takes much longer to complete. It is intended for hard, high-stakes problems where the extra accuracy justifies the cost. For everyday coding, agentic, and chat workloads, use GPT-6 Luna instead.

Learn more in OpenAI's docs: https://developers.openai.com/api/docs/guides/reasoning#reasoning-mode

Text1.1M context$0.10 / $0.50

GPT-6 Luna Pro (batch)

GPT-6 Luna Pro is the same underlying model as GPT-6 Luna, served with reasoning.mode set to pro for higher-quality responses on complex tasks.

Learn more in OpenAI's docs: https://developers.openai.com/api/docs/guides/reasoning#reasoning-mode

Text1.1M context$0.05 / $0.25

GPT-6 Luna

GPT-6 Luna is the fast, cost-efficient model in OpenAI's GPT-6 series, positioned below GPT-6 Sol. It is suited for high-volume and latency-sensitive workloads such as chat, classification, and lightweight agentic tasks, and at higher reasoning effort it can take on complex software engineering and computer-use tasks that previously called for a Sol-tier model. It shares the GPT-6 family's gains in factual reliability and its clearer, more concise communication style.

Text1.1M context$0.10 / $0.50

GPT-6 Luna (batch)

Text1.1M context$0.05 / $0.25

GPT-6 Sol Pro

GPT-6 Sol Pro is the same underlying model as GPT-6 Sol, served with reasoning.mode set to pro for higher-quality responses on complex tasks.

Cost note: pro mode spends far more reasoning tokens per request, so a typical request costs several times more than the same request on GPT-6 Sol and takes much longer to complete. It is intended for hard, high-stakes problems where the extra accuracy justifies the cost. For everyday coding, agentic, and chat workloads, use GPT-6 Sol instead.

Learn more in OpenAI's docs: https://developers.openai.com/api/docs/guides/reasoning#reasoning-mode

Text1.1M context$2 / $10

GPT-6 Sol Pro (batch)

GPT-6 Sol Pro is the same underlying model as GPT-6 Sol, served with reasoning.mode set to pro for higher-quality responses on complex tasks.

Learn more in OpenAI's docs: https://developers.openai.com/api/docs/guides/reasoning#reasoning-mode

Text1.1M context$1 / $5

GPT-6 Sol

GPT-6 Sol is the cost-efficient high-end model in OpenAI's GPT-6 series, positioned below the flagship GPT-6 Astra and above the fast GPT-6 Luna tier. It is suited for demanding professional work, agentic coding, business workflow automation, and computer use, and is particularly strong at long-horizon software engineering tasks in real codebases. It approaches Astra-level factual reliability at a much lower cost and shares Astra's clearer, more concise communication style in technical and coding conversations.

Text1.1M context$2 / $10

GPT-6 Sol (batch)

Text1.1M context$1 / $5

GPT Astra Latest

This model always redirects to the latest model in the GPT Astra family.

Text1.1M context

OpenAI: GPT-5.6 Luna

Providers