OpenAILLM matchup

GPT-6 Lunavs.GPT-5.6 Terra

Is the higher price worth it?

GPT-5.6 Terra performs better than GPT-6 Luna, but costs 20 times (input) and 24 times (output) as much.

GPT-5.6 Terra ahead

IntelligenceOverall index, 100 points = average

96.3

GPT-6 Luna

Rank 35

100.7

GPT-5.6 Terra

Rank 26

about as fast

SpeedTokens per second

101

GPT-6 Luna

100

GPT-5.6 Terra

GPT-6 Luna cheaper

Price$ per 1M tokens

$0.20

GPT-6 Luna

$4.50

GPT-5.6 Terra

about as popular

Popularity0 to 100

GPT-6 Luna

Rank 30 of 785

GPT-5.6 Terra

Rank 19 of 785

Above 272,000 tokens of context, GPT-6 Luna costs more: $0.20 input and $0.75 output per 1M tokens. Above 272,000 tokens of context, GPT-5.6 Terra costs more: $4 input and $18 output per 1M tokens.

Choose GPT-6 Luna for

No clear advantage

Choose GPT-5.6 Terra for

LegalEducationProductivityMultimodalCodingReasoningGeneral (not certain)Finance (not certain)

Where is the difference?

Strengths at a glance

GPT-6 LunaGPT-5.6 TerraMedian of all ranked models

CodingGPT-5.6 Terra ahead

GPT-6 Luna97.5Rank 35

GPT-5.6 Terra104.0Rank 21

Median of ranked models84.1

95 % range: GPT-6 Luna 94.5 to 100.5, GPT-5.6 Terra 98.6 to 109.3.

AreaGPT-6 LunaGPT-5.6 Terra

The same values and verdicts as the bars below. In the close-up the net starts at 80 points, just below the weaker model's lowest value; the reference group average is 100. Not in the net because a model has no own value there or the data are too thin: Mathematics.

GPT-5.6 Terra is ahead in 4 of 10 areas, GPT-6 Luna in none.

level: 3too little data: 1GPT-5.6 Terra slightly ahead: 2GPT-5.6 Terra ahead: 4

Areas in the overall index

MultimodalGPT-5.6 Terra +792.0 to 99.4

CodingGPT-5.6 Terra +697.5 to 104.0

ReasoningGPT-5.6 Terra +697.8 to 103.6

Generalleaning GPT-5.6 Terranot certain · 5 and 4 tests

Knowledgelevel89.2 to 91.1

Agentslevel99.5 to 100.9

Mathematicstoo little data105.5 to 101.7

Specialist fields

LegalGPT-5.6 Terra +1190.7 to 101.6

Financeleaning GPT-5.6 Terranot certain · 5 and 6 tests

Healthlevel95.2 to 96.6

Bar length shows the gap in index points (full width 15 points, one standard deviation of the reference group). Solid means clearly ahead, hatched not certain yet, a thin grey bar level.

Show 4 more areas that do not count toward the overall index

EducationGPT-5.6 Terra +993.0 to 102.0

ProductivityGPT-5.6 Terra +890.5 to 98.8

Cybersecuritytoo little data101.8 to 104.2

Searchonly GPT-6 Luna has a score

Bar length shows the gap in index points (full width 15 points, one standard deviation of the reference group). Solid means clearly ahead, hatched not certain yet, a thin grey bar level.

GPT-6 Luna vs. GPT-5.6 Terra at the same effort

Both models measured at the same reasoning efforts. This shows whether a lead comes from the model or only from more compute.

maxGPT-5.6 Terra ahead

GPT-6 Luna 96.3 vs. GPT-5.6 Terra 100.7 · Gap 4.5 · Test series 11

Overall index (max vs. max)GPT-5.6 Terra ahead

GPT-6 Luna 96.3 vs. GPT-5.6 Terra 100.7

Effort	GPT-6 Luna	GPT-5.6 Terra	Gap	Verdict	Test series
max	96.3	100.7	4.5	GPT-5.6 Terra ahead	11
Overall index (max vs. max)	96.3	100.7	4.5	GPT-5.6 Terra ahead

Gap and verdict compare only the Artificial Analysis test series in which both models are measured at that effort. Tied below 3 points, ahead only beyond two standard errors. The single values rest on all test series of each model, so their difference can deviate from the gap.

Which model for which task?

Everyday tasks, translated from the area values above. Each row names the values it rests on.

TaskBetter choicebased on

Writing and debugging codeGPT-5.6 TerraCoding 97.5 to 104.0

Maths and data analysisGPT-5.6 TerraMathematics 105.5 to 101.7 (too little data), Reasoning 97.8 to 103.6

Facts and researchlevelKnowledge 89.2 to 91.1, Search without a score for GPT-5.6 Terra

Agents and workflowslevelAgents 99.5 to 100.9

Fast chat repliesabout the same101 to 100 tokens per second

Security analysisopenCybersecurity 101.8 to 104.2 (too little data)

Many requests on a small budgetGPT-6 Luna$0.20 to $4.50 per 1M tokens

Finance, legal and medical textsGPT-5.6 TerraFinance 98.3 to 101.7, Legal 90.7 to 101.6, Health 95.2 to 96.6

Images, screenshots and PDFsGPT-5.6 TerraMultimodal 92.0 to 99.4

See for yourself: same task, one attempt

The same prompt for both models. We show the latest attempt without fixes; it does not count toward the overall index.

Pelican on a bicycle

Prompt

Generate an SVG of a pelican riding a bicycle

Task origin: Simon Willison

GPT-6 Luna

Oct 1, 2026 · 47 s · $0.0022

Test details

Run date: Oct 1, 2026
API provider: openrouter
Requested API model: openai/gpt-6-luna
Reasoning setting: high
Token limit including reasoning: 8,192 tokens
Output tokens: 4,493 tokens
Cost of the attempt: $0.0022
Response time: 46.7 s
API model reported in response: openai/gpt-6-luna
Pinned endpoint: OpenAI | openai/gpt-6-luna-20260922
Temperature: Provider default
Test protocol: community-visual-tests-v1
Technical checks: Passed, not a quality rating

Visually reviewed. This individual test does not establish overall model quality.

GPT-5.6 Terra

Sep 7, 2026 · 65 s · $0.04

Test details

Run date: Sep 7, 2026
API provider: openrouter
Requested API model: openai/gpt-5.6-terra
Reasoning setting: high
Token limit including reasoning: 8,192 tokens
Output tokens: 3,373 tokens
Cost of the attempt: $0.0405
Response time: 64.6 s
API model reported in response: openai/gpt-5.6-terra
Pinned endpoint: OpenAI | openai/gpt-5.6-terra-20260709
Temperature: Provider default
Test protocol: community-visual-tests-v1
Technical checks: Passed, not a quality rating

Visually reviewed. This individual test does not establish overall model quality.

SVG short test

One SVG task with the same fixed settings for every model, without hidden thinking.

Prompt

Return only one standalone SVG of a pelican riding a bicycle. Use viewBox="0 0 800 600", simple shapes, two clearly visible wheels and a recognizable pelican with a long beak. At most 30 SVG elements and 1800 characters. No Markdown, explanations, comments, scripts, animation, CSS or external resources.

GPT-6 Luna

Unavailable in this round

GPT-5.6 Terra

Unavailable in this round

GPT-6 Luna and GPT-5.6 Terra in the overall index

Every model with a place in the overall index, both models of this matchup highlighted. You can add or hide other models.

Overall index (Gradually Meta-Score v4)

20 of 92 models

4 models in this view are measured at a lower reasoning effort than max in the overall index. Dashed column: lower reasoning effort than the top.

Claude Opus 5.5max126.1

Gemini 4 Carbonhigh125.8

Claude Fable 5.1max123.8

Claude Sonnet 5.5max122.8

Gemini 4 Argonhigh122.3

GPT-6 Astramax121.3

GPT-6.1 Solmax119.2

Muse Wonder Fullmax118.3

Claude Opus 5max117.2

Muse Wondermax113.4

Kimi K3.1 Promax112.9

Claude Fable 5max112.5

GPT-6 Solmax110.5

GPT-5.6 Solmax108.7

Muse Spark 1.3max107.4

Gemini 3.7 Flashhigh107.1

DeepSeek V4.1 Flash 1017max106.6

Gemini 3.8 Flashhigh106.5

GPT-5.6 Terramax100.7

GPT-6 Lunamax96.3

Claude Opus 5.5

Gemini 4 Carbon

Claude Fable 5.1

Claude Sonnet 5.5

Gemini 4 Argon

GPT-6 Astra

GPT-6.1 Sol

Muse Wonder Full

Claude Opus 5

Muse Wonder

Kimi K3.1 Pro

Claude Fable 5

GPT-6 Sol

GPT-5.6 Sol

Muse Spark 1.3

Gemini 3.7 Flash

DeepSeek V4.1 Flash 1017

Gemini 3.8 Flash

GPT-5.6 Terra

GPT-6 Luna

Popularity compared

How often the models are searched for, used and mentioned, on a scale from 0 to 100.

Both are about as popular: GPT-6 Luna 60, GPT-5.6 Terra 64 out of 100.

Over time

Weekly values from October 18, 2026 to November 7, 2026

GPT-6 LunaGPT-5.6 Terra

All data side by side

The better value is bold. For the overall index only when the gap is clear.

Metric	GPT-6 Luna	GPT-5.6 Terra
Intelligence
Overall indexpoints, 100 = reference group average, higher is better	96.3	100.7
Rank	Rank 35 of 92	Rank 26 of 92
Price per 1M tokens
Inputwhat you send	$0.10above 272,000 tokens: $0.20	$2above 272,000 tokens: $4
Outputwhat the model writes	$0.50above 272,000 tokens: $0.75	$12above 272,000 tokens: $18
Cost per task (own tests)average of 3 original-protocol tasks both models passed (SVG short test not included)	$0.003	$0.047
Speed
Tokens per second	101	100
Popularity
Popularitysearch, usage and mentions, 0 to 100	60Rank 30 of 785	64Rank 19 of 785
Technical
Context windowhow much text the model reads at once	1.05 M tokens≈ 1,580 A4 pages	1.05 M tokens≈ 1,580 A4 pages
Knowledge cutoff	May 18, 2026	Feb 16, 2026
Released	Sep 22, 2026	Jul 9, 2026
Availability	Active	Active
Open weights	no	no

Prices without caching or batch discounts · Page count as a rule of thumb from Artificial Analysis

Is there a better choice?

Successor of GPT-5.6 TerraGPT-6.1 Terra

Overall index 105.3, rank 19$1 input and $5 output per 1M tokens

GPT-6 Luna vs. GPT-6.1 Terra

As cheap as GPT-6 LunaClaude Haiku 5.5

Overall index 103.4, rank 20$0.10 input and $0.50 output per 1M tokens

GPT-5.6 Terra vs. Claude Haiku 5.5

Best model in the overall indexClaude Opus 5.5

Overall index 126.1, rank 1$4 input and $20 output per 1M tokens

GPT-5.6 Terra vs. Claude Opus 5.5

Shared benchmarks

Single results from the same benchmark version, task and metric. A value is bold only with the same test setup and a clear gap.

A: GPT-6 LunaB: GPT-5.6 Terra

AA-BriefcaseOverall · higher is better1,337.1 elo1,333.1 eloArtificial AnalysisAs of 11/07/2026

AA-OmniscienceOverall · higher is better-8.267 points-9.267 pointsArtificial AnalysisAs of 11/07/2026

AA-Omniscience AccuracyOverall · higher is better39.6%42.3%Artificial AnalysisAs of 11/07/2026

AA-Omniscience Hallucination RateOverall · lower is better79.3%89.4%Artificial AnalysisAs of 11/07/2026

AutomationBench-AAOverall · higher is better53.2%59.6%Artificial AnalysisAs of 11/07/2026

Code MigrationOverall · higher is better42.6%47.8%Vals AIAs of 10/29/2026

CritPt v2 (Hard Subset)Overall · higher is better42.5%50.5%Artificial AnalysisAs of 11/07/2026

CyberBench v1.1Overall · higher is better76.3%72.1%Vals AIAs of 11/05/2026

EMBOverall · higher is better68.5%66.2%Vals AIAs of 10/20/2026

Finance Agent (v2)Overall · higher is better49.9%54.4%Vals AIAs of 10/29/2026

Show 54 more benchmarks

FrontierCodeOverall · higher is better25.2%23.9%Artificial AnalysisAs of 11/07/2026

GDP.pdfOverall · higher is better22.8%24.0%Artificial AnalysisAs of 11/07/2026

GDPval-AAOverall · higher is better1,444.5 elo1,461.5 eloArtificial AnalysisAs of 11/07/2026

HLE-DiamondOverall · higher is better22.2%28.8%Artificial AnalysisAs of 11/07/2026

IOIOverall · higher is better55.6%87.6%Vals AIAs of 10/29/2026

Legal Research BenchOverall · All-pass · higher is better30.3%41.3%Vals AIAs of 11/05/2026

LMArena Text, Style ControlOverall · higher is betterTest setup not documented as identical1,4411,467LMArenaAs of 10/28/2026

LMArena Vision, Style ControlOverall · higher is betterTest setup not documented as identical1,2401,268LMArenaAs of 10/19/2026

MedCodeOverall · higher is betterTest setup not documented as identical44.7%43.4%Vals AIAs of 10/29/2026

MedScribeOverall · higher is betterTest setup not documented as identical83.7%82.9%Vals AIAs of 10/29/2026

MMMU-ProOverall · higher is better79.7%80.7%Artificial AnalysisAs of 11/07/2026

ProgramBenchOverall · higher is better0.5%0.5%Vals AIAs of 11/05/2026

ProofBench v1.1Overall · higher is better64%74%Vals AIAs of 10/29/2026

Public Benefits Bench v1.1Overall · higher is better57.6%62.4%Vals AIAs of 10/12/2026

SAGEOverall · higher is better83.4%87.0%Vals AIAs of 11/05/2026

SRE BenchOverall · higher is better2.7%9.5%Vals AIAs of 11/05/2026

Tax Agent BenchOverall · higher is better19.9%30.0%Vals AIAs of 11/05/2026

Terminal-BenchOverall · higher is better12.6%35.4%Artificial AnalysisAs of 11/07/2026

Terminal-Bench 4.0Overall · higher is better13.6%22.7%Vals AIAs of 11/05/2026

Terminal-Bench-ScienceOverall · higher is better4.3%10.0%Vals AIAs of 11/05/2026

Terminal-Bench-ScienceOverall · higher is better8.6%13.8%Artificial AnalysisAs of 11/07/2026

Vibe Code Bench v1.1Overall · higher is better81.6%74.6%Vals AIAs of 11/05/2026

Code MigrationCLI · higher is better37.2%40.4%Vals AIAs of 10/29/2026

CyberBench v1.1Patch · higher is better87.5%87.5%Vals AIAs of 11/05/2026

EMBComps · higher is better70.5%62.2%Vals AIAs of 10/20/2026

Finance Agent (v2)Adjustments · higher is better49.8%49.7%Vals AIAs of 10/29/2026

IOIIOI 2024 · higher is better69.3%85.0%Vals AIAs of 10/29/2026

Legal Research BenchAdministrative / Regulatory · All-pass · higher is better37.9%57.6%Vals AIAs of 11/05/2026

LMArena Text, Style ControlBusiness, management and finance · higher is betterTest setup not documented as identical1,4411,466LMArenaAs of 10/28/2026

LMArena Vision, Style ControlChinese · higher is betterTest setup not documented as identical1,2921,313LMArenaAs of 10/19/2026

ProgramBenchAlmost Resolved · higher is better8%15%Vals AIAs of 11/05/2026

SAGEComputer Science · higher is better86.6%89.2%Vals AIAs of 11/05/2026

SRE BenchCapability Score · higher is better11.1%26.8%Vals AIAs of 11/05/2026

Tax Agent BenchControversy & Precedence Analysis · higher is better6.7%16.7%Vals AIAs of 11/05/2026

Terminal-Bench 4.0Hardware · higher is better0%0%Vals AIAs of 11/05/2026

Terminal-Bench-ScienceEarth Sciences · higher is better0%0%Vals AIAs of 11/05/2026

Code MigrationCOBOL · higher is better58.6%70.0%Vals AIAs of 10/29/2026

CyberBench v1.1PoC · higher is better65.0%56.7%Vals AIAs of 11/05/2026

EMBDataroom · higher is better75.5%82.2%Vals AIAs of 10/20/2026

Finance Agent (v2)All-Pass · higher is better38.4%42.8%Vals AIAs of 10/29/2026

IOIIOI 2025 · higher is better39.0%77.8%Vals AIAs of 10/29/2026

Legal Research BenchAdministrative / Regulatory · Weighted partial credit (weighted pass) · higher is better83.2%88.8%Vals AIAs of 11/05/2026

LMArena Text, Style ControlChinese · higher is betterTest setup not documented as identical1,4831,516LMArenaAs of 10/28/2026

LMArena Vision, Style ControlCreative writing with images · higher is betterTest setup not documented as identical1,2321,254LMArenaAs of 10/19/2026

ProgramBenchFully Resolved · higher is better0.5%0.5%Vals AIAs of 11/05/2026

SAGELinear Algebra · higher is better76.5%79.5%Vals AIAs of 11/05/2026

SRE BenchFully Solved · higher is better2.7%9.5%Vals AIAs of 11/05/2026

Tax Agent BenchControversy & Precedence Analysis (weighted pass) · higher is better54.4%64.8%Vals AIAs of 11/05/2026

Terminal-Bench 4.0Media · higher is better16.7%16.7%Vals AIAs of 11/05/2026

Terminal-Bench-ScienceEngineering Sciences · higher is better11.1%0.0%Vals AIAs of 11/05/2026

Code MigrationCode Quality · higher is better3.752 / 53.948 / 5Vals AIAs of 10/29/2026

EMBDCF · higher is better62.8%53.4%Vals AIAs of 10/20/2026

Finance Agent (v2)Comparables · higher is better39.8%47.0%Vals AIAs of 10/29/2026

IOIIOI 2026 · higher is better58.3%100.0%Vals AIAs of 10/29/2026

More comparisons

Compare GPT-6 Luna and GPT-5.6 Terra with the models people compare them with most often.

GPT-6 Lunavs.GPT-5.6 Terra

Choose GPT-6 Luna for

Choose GPT-5.6 Terra for

Where is the difference?

Strengths at a glance

GPT-6 Luna vs. GPT-5.6 Terra at the same effort

Which model for which task?

See for yourself: same task, one attempt

Pelican on a bicycle

SVG short test

GPT-6 Luna

GPT-5.6 Terra

GPT-6 Luna and GPT-5.6 Terra in the overall index

Overall index (Gradually Meta-Score v4)

Popularity compared

Over time

All data side by side

Is there a better choice?

Shared benchmarks

More comparisons

Compare another matchup

Predecessor and successor

The most popular comparisons