Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models838Deals30d267Policy30d79, 15 pendingSources1,184

GPT-6 Luna

OpenAIReleased Sep 22, 2026Current

Score basis

GPT-6 Luna is strong in long context; capable in reasoning and multimodal tasks; and behind the leaders in factuality, math, agentic tasks, and coding. Too few results yet to rate safety, multilingual tasks, or instruction following.

Capability profile

Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.

Strong1 capability
  1. Long Context−8.3%1 of 3
Capable2 capabilities
  1. Reasoning−23.7%4 of 6
  2. Multimodal−24.9%1 of 6
Limited4 capabilities
  1. Factuality−25.4%3 of 4
  2. Math−29.4%3 of 5
  3. Agentic−32.0%3 of 7
  4. Coding−32.9%4 of 10
Not rated3 capabilities

Too few results yet to rate Safety, Multilingual or Instruction Following.

Price

$0.10input$0.50outputper million tokens

From Artificial Analysis · 3 providers tracked · All prices

Cheaper than 79% of 331 priced models · 3:1 input-to-output blend, log scale

Evidence

139results on44benchmarks

  • 37 independently verified
  • 89 aggregator
  • 7 vendor-reported
  • 6 cross-referenced

From 15 sources · latest Nov 13, 2026 · How verification works

API features

Tool callingStructured outputsReasoning

As listed by OpenRouter

Research

26 papers reference GPT-6 Luna

GPT-6 Luna benchmark results

139 results on 44 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Long Context

StrongFull long context ranking

8.3% behind the leader1 of 3 ranked benchmarks measured

Show 5 more long context resultsHide 5 long context results

Reasoning

CapableFull reasoning ranking

23.7% behind the leader4 of 6 ranked benchmarks measured

Show 28 more reasoning resultsHide 28 reasoning results

Multimodal

CapableFull multimodal ranking

24.9% behind the leader1 of 6 ranked benchmarks measured

Show 6 more multimodal resultsHide 6 multimodal results

Factuality

LimitedFull factuality ranking

25.4% behind the leader3 of 4 ranked benchmarks measured

Show 10 more factuality resultsHide 10 factuality results

Math

LimitedFull math ranking

29.4% behind the leader3 of 5 ranked benchmarks measured

Agentic

LimitedFull agentic ranking

32.0% behind the leader3 of 7 ranked benchmarks measured

Show 10 more agentic resultsHide 10 agentic results

Coding

LimitedFull coding ranking

32.9% behind the leader4 of 10 ranked benchmarks measured

Show 5 more coding resultsHide 5 coding results

Instruction Following

Not enough dataFull instruction following ranking

0 of 3 ranked benchmarks measured

More results

Benchmarks outside the capability baskets. They are not ranked against a leader.

Show 50 more resultsHide 50 results

GPT-6 Luna: common questions

Who makes GPT-6 Luna?

GPT-6 Luna is made by OpenAI.

When was GPT-6 Luna released?

GPT-6 Luna was released on Sep 22, 2026, according to Artificial Analysis.

What is GPT-6 Luna good at?

GPT-6 Luna is strong in long context; capable in reasoning and multimodal tasks; and behind the leaders in factuality, math, agentic tasks, and coding. Too few results yet to rate safety, multilingual tasks, or instruction following.

How much does GPT-6 Luna cost?

GPT-6 Luna costs $0.10 per million input tokens and $0.50 per million output tokens, according to Artificial Analysis. We track its price at 3 providers. At a mix of three input tokens to one output token, it is cheaper than 79% of the 331 priced models we track.

How many benchmarks has GPT-6 Luna been tested on?

We track 139 results for GPT-6 Luna on 44 benchmarks from 15 sources, 37 of them independently verified. The latest was recorded on Nov 13, 2026.

Which API features does GPT-6 Luna support?

OpenRouter lists tool calling, structured outputs, and reasoning for GPT-6 Luna.

About this record

Where GPT-6 Luna's numbers come from, and every name it appears under.

Tracked since
Sep 22, 2026
Newest source mention
Nov 14, 2026

Where the results come from

Verification: 139 scores · 37 independently verified · 89 aggregator-attributed · 6 vendor cross-reference · 7 vendor-reported. How these tiers are assigned

From 15 sources on 9 sites. Artificial Analysis supplies 90 of them; the 37 independently verified results come from 5 sites. Bars are coloured by trust tier.

  • artificialanalysis.ai90
  • arcprize.org24
  • livebench.ai7
  • anthropic.com4
  • deploymentsafety.openai.com4
  • api.llm-stats.com3
  • epoch.ai3
  • assets.claude.com2
  • datasets-server.huggingface.co2

Also known as

How our sources name GPT-6 Luna at each reasoning setting.

SettingShort formAPI id
lowgpt-6 luna (low)gpt-6-luna-low
mediumgpt-6 luna (medium)gpt-6-luna-medium
highgpt-6 luna (high) gpt-6 luna - provider adapter (high)gpt-6-luna-high
xhighgpt-6 luna (xhigh) gpt-6 luna - provider adapter (xhigh)gpt-6-luna-xhigh
maxgpt-6 luna (max) gpt-6 luna - provider adapter (max)gpt-6-luna-max
Also listed asgpt-6-luna-non-reasoninggpt-6 luna (non-reasoning)gpt-6 luna (none)gpt-6-luna:batchgpt6-luna