Compare / GPT-5.6 Luna vs GPT-6 Luna

GPT-5.6 Luna vs GPT-6 Luna

Which is cheaper, which is more capable, and which does better at each kind of work, from 18 results on 4 sources that measured both models.

OpenAIGPT-5.6 Luna Released 9 Jul 2026 · $0.20 in · $1.20 out
OpenAIGPT-6 Luna Released 22 Sep 2026 · $0.10 in · $0.50 out
GPT-5.6 Luna wins 48 too close to callGPT-6 Luna wins 6

Of the 18 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 16 November 2026 (catalogue release 2026-11-16-49bdfa045dee).

GPT-5.6 Luna or GPT-6 Luna? The short answer

GPT-6 Luna is the newer version; they score about the same, so pick GPT-6 Luna, which costs less (2.2 times less per token).

  • GPT-6 Luna is cheaper: $0.20 against $0.45 per million tokens, 2.2 times less.
  • They score about the same on the Artificial Analysis Intelligence Index (37.3 and 38.1).
  • GPT-6 Luna is newer: released 22 Sep 2026, against 9 Jul 2026 for GPT-5.6 Luna.
  • On third-party benchmarks, on the results both models have: GPT-6 Luna does better for agents and tool use and professional work.

Head to head

37.3
Intelligence Index
38.1
$0.45
Price per million tokens, blendedlower is better
$0.20
1,050,000
Context window, tokens
1,050,000

Every fact side by side

GPT-5.6 LunaGPT-6 Luna
Price per million tokens$0.20 in · $1.20 out$0.10 in · $0.50 out
Blended (3 in : 1 out)$0.45$0.20
Cheapest host, blended$0.45 (Amazon Bedrock)$0.20 (Amazon Bedrock)
Cached input, per million$0.020 cached input$0.010 cached input
Intelligence Index37.338.1
Spring Prompt overall–39
Output speed, tokens a second–127
Context window1,050,000 tokens1,050,000 tokens
Released9 Jul 202622 Sep 2026 (newer)
WeightsClosedClosed
Developer based inthe USthe US

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.

Product listingsTurning a sparse product feed and photos into listings that can go live not measured 7 of 32
Decks from an analysisTurning a finished analysis into a deck you could present as it is not measured 10 of 32
User surveysPlanning a user survey and reading its results without being misled not measured 12 of 23
Marketing planningPlanning a year of ad spend without overspending not measured 6 of 32
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 58 of 159 53 of 159
Professional workReal tasks from banking, consulting and law, business documents and freelance projects 20 of 51 25 of 51
Reasoning and knowledgeHard questions across science, maths and general knowledge 44 of 187 48 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality 60 of 169 61 of 169
Sticking to the factsSummarising without inventing things, and factual answers 60 of 139 89 of 139
Speed under pressureGood decisions against a real clock (fast chess) not measured 24 of 35

From our research

Every shared result

Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.

MeasureGPT-5.6 LunaGPT-6 Luna
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Overall · TextArena rating, higher is better 64 · 78 of 387 measured 1,453xhigh reasoning 1,443max reasoning
Overall · Text factualityArena rating, higher is better 40 · 98 of 163 measured 1,460xhigh reasoning 1,430max reasoning
Business, management and finance · TextArena rating, higher is better 47 · 71 of 380 measured 1,463xhigh reasoning 1,447max reasoning
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 41 · 39 of 249 measured 37.3max reasoning 38.1max reasoning
Humanity's Last Exam% of questions, higher is better 49 · 53 of 250 measured 39.5%max reasoning 38.5%max reasoning
Terminal-Bench 4.0% of tasks, higher is better 42 · 37 of 120 measured 11.6%max reasoning 12.6%max reasoning
GDP.pdf reported by GDP.pdf · 1 result
Professional document tasks% of rubric, higher is better 15 · 14 of 32 measured 22.7% 23.0%max reasoning
SimpleQA Verified (Epoch AI) reported by SimpleQA Verified (Epoch AI) · 1 result
Correct answers% of questions, higher is better 42 · 40 of 76 measured 41.0%max reasoning 41.4%max reasoning
Show all 18 results (10 more)
MeasureGPT-5.6 LunaGPT-6 Luna
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 8 results
Confirmed task success · AgentIPS effect estimate, higher is better 34 · 29 of 48 measured -0.03xhigh reasoning, arena agent -0.00max reasoning, arena agent
Creative writing · TextArena rating, higher is better 80 · 77 of 385 measured 1,409xhigh reasoning 1,413max reasoning
Expert prompts · TextArena rating, higher is better 51 · 40 of 338 measured 1,491xhigh reasoning 1,498max reasoning
Instruction following · TextArena rating, higher is better 63 · 57 of 387 measured 1,445xhigh reasoning 1,448max reasoning
Praise over complaint · AgentIPS effect estimate, higher is better 35 · 30 of 48 measured -0.04xhigh reasoning, arena agent -0.02max reasoning, arena agent
Steerability · AgentIPS effect estimate, higher is better 31 · 16 of 48 measured 0.00xhigh reasoning, arena agent 0.04max reasoning, arena agent
Tool grounding · AgentIPS effect estimate, higher is better 22 · 39 of 48 measured 0.00xhigh reasoning, arena agent -0.00max reasoning, arena agent
Writing, literature and language · TextArena rating, higher is better 64 · 79 of 386 measured 1,433xhigh reasoning 1,421max reasoning
Artificial Analysis reported by Artificial Analysis · 2 results
Long-context reasoning (AA-LCR)% of questions, higher is better 11 · 13 of 241 measured 83.7%max reasoning 83.3%max reasoning
SciCode% of problems, higher is better 37 · 31 of 121 measured 53.6%max reasoning 54.6%max reasoning

Quick answers

Is GPT-5.6 Luna better than GPT-6 Luna?

GPT-6 Luna is the newer version; they score about the same, so pick GPT-6 Luna, which costs less (2.2 times less per token).

Which is cheaper, GPT-5.6 Luna or GPT-6 Luna?

GPT-6 Luna is cheaper: $0.20 against $0.45 per million tokens, 2.2 times less.

Which is better for agents and tool use, GPT-5.6 Luna or GPT-6 Luna?

GPT-6 Luna ranks 53 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 58 for GPT-5.6 Luna.

Which is better for professional work, GPT-5.6 Luna or GPT-6 Luna?

GPT-5.6 Luna ranks 20 of 51 models on sale for real tasks from banking, consulting and law, business documents and freelance projects, against 25 for GPT-6 Luna.

Which is better for reasoning and knowledge, GPT-5.6 Luna or GPT-6 Luna?

GPT-5.6 Luna ranks 44 of 187 models on sale for hard questions across science, maths and general knowledge, against 48 for GPT-6 Luna.

Which is better for writing, GPT-5.6 Luna or GPT-6 Luna?

GPT-5.6 Luna ranks 60 of 169 models on sale for what people prefer in blind comparisons, and judged writing quality, against 61 for GPT-6 Luna.

Which is better for sticking to the facts, GPT-5.6 Luna or GPT-6 Luna?

GPT-5.6 Luna ranks 60 of 139 models on sale for summarising without inventing things, and factual answers, against 89 for GPT-6 Luna.

Run GPT-5.6 Luna and GPT-6 Luna on your own prompt.

Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.