Enter opens the highlighted result. Esc closes.

GPT-5.6 Terra vs GPT-6 Luna

Updated October 21, 2026. 95 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace GPT-5.6 Terra Replace GPT-6 Luna
GPT-5.6 TerraOpenAI
Coding
69.0
Overall
42.1
Output $/M
$12.00
72
GPT-6 LunaOpenAI
Coding
71.4
Overall
38.1
Output $/M
$0.50
22
vs boards won of 95 shared, 1 tied

Decision reading

Pick GPT-6 Luna for coding: it scores 71.4 on the Vibe Coding Index against 69.0 for GPT-5.6 Terra.

Pick GPT-5.6 Terra when what matters is LiveBench coding or Terminal-Bench 4 (Vals AI), where it leads.

Model AGPT-5.6 TerraOpenAI69.0/1002 independent benchmarks$0.42 per solved taskModel BGPT-6 LunaOpenAI71.4/1002 independent benchmarksNo cost per solved task
Shared results
95
GPT-5.6 Terra only
29
GPT-6 Luna only
7
Like-for-like categories
8 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
GPT-5.6 Terra 69.0 from 2 benchmarks, GPT-6 Luna 71.4 from 2 benchmarks.
Widest category
Reasoning, 11.2 points apart on average over 12 shared benchmarks.
Widest single gap
IOI (Vals AI): GPT-5.6 Terra 87.6% against GPT-6 Luna 55.6%, 32.1 points apart on a 0 to 100 scale.
Output price
GPT-5.6 Terra $12.00 against GPT-6 Luna $0.50 per million output tokens, 24.0x apart.
Context window
Both take 1.05M tokens.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryGPT-5.6 TerraBarsGPT-6 LunaTies
Coding971 tie
Intelligence20
Agentic work94
Reasoning120
Math60
Knowledge51
Web design27
General273
All shared boards72221 tie
CodingLike-for-likeFour of these boards feed the Vibe Coding Index97
Coding benchmarks for GPT-5.6 Terra and GPT-6 Luna
BenchmarkGPT-5.6 TerraGPT-6 LunaLeader
LiveBench coding86.2%84.5%GPT-5.6 Terra
LMArena WebDev15221578GPT-6 Luna
AA Coding Index76.7Not measuredOnly GPT-5.6 Terra measured
LiveCodeBench (Vals AI)85.9%Not measuredOnly GPT-5.6 Terra measured
SWE-bench Verified (Vals AI)95.4%Not measuredOnly GPT-5.6 Terra measured
Terminal-Bench 4 (Vals AI)22.7%13.6%GPT-5.6 Terra
Vibe Code Bench (Vals AI)74.6%81.7%GPT-6 Luna
Code migration (Vals AI)47.8%42.5%GPT-5.6 Terra
ProgramBench (Vals AI)0.5%0.5%Tie
IOI (Vals AI)87.6%55.6%GPT-5.6 Terra
Terminal-Bench 2.1 (Vals AI)77.5%73.0%GPT-5.6 Terra
SWE-bench Multimodal (Vals AI)Not measured37.5%Only GPT-6 Luna measured
Vibe Code Bench 1-100 (Vals AI)14.8%Not measuredOnly GPT-5.6 Terra measured
SkillsBench (Vals AI)58.9%Not measuredOnly GPT-5.6 Terra measured
CyberBench (Vals AI)72.1%76.3%GPT-6 Luna
SciCode55.0%54.6%GPT-5.6 Terra
Terminal-Bench Hard (Artificial Analysis run)62.9%Not measuredOnly GPT-5.6 Terra measured
FrontierCode41.3%42.4%GPT-6 Luna
WeirdML78.3%Not measuredOnly GPT-5.6 Terra measured
ALE-Bench1951.41576.9GPT-5.6 Terra
LMArena Text: coding14851437GPT-5.6 Terra
LMArena Text: software and IT14821423GPT-5.6 Terra
LMArena WebDev: React15131573GPT-6 Luna
LMArena WebDev: HTML15261595GPT-6 Luna
LMArena WebDev: from an image15411571GPT-6 Luna
Neither model has a result on the other 13 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for GPT-5.6 Terra and GPT-6 Luna
BenchmarkGPT-5.6 TerraGPT-6 LunaLeader
AA Intelligence Index42.138.1GPT-5.6 Terra
Epoch Capabilities Index160.1156.8GPT-5.6 Terra
Agentic workLike-for-likePublic boards, shown as published, kept out of the index94
Agentic work benchmarks for GPT-5.6 Terra and GPT-6 Luna
BenchmarkGPT-5.6 TerraGPT-6 LunaLeader
AA Agentic Index43.2Not measuredOnly GPT-5.6 Terra measured
Design Arena: agents building web apps11201189GPT-6 Luna
Design Arena: agents building full-stack apps10611158GPT-6 Luna
Design Arena: agents building mobile apps11121133GPT-6 Luna
Design Arena: agents building Android apps1066Not measuredOnly GPT-5.6 Terra measured
Design Arena: agents building games1082Not measuredOnly GPT-5.6 Terra measured
Design Arena: HTML slides1108Not measuredOnly GPT-5.6 Terra measured
Design Arena: PowerPoint slides1130Not measuredOnly GPT-5.6 Terra measured
LiveBench research60.4%43.5%GPT-5.6 Terra
LiveBench browser84.0%72.0%GPT-5.6 Terra
LiveBench long agentic48.4%48.1%GPT-5.6 Terra
SREBench (Vals AI)9.5%2.7%GPT-5.6 Terra
Finance Agent (Vals AI)54.4%49.9%GPT-5.6 Terra
Tax Agent Bench (Vals AI)30.0%19.9%GPT-5.6 Terra
Legal Research Bench (Vals AI)41.4%30.3%GPT-5.6 Terra
Harvey's Legal Agent Benchmark (Vals AI)Not measured4.2%Only GPT-6 Luna measured
Excel Modeling Benchmark (Vals AI)66.2%68.5%GPT-6 Luna
BioMysteryBench (Vals AI)Not measured61.5%Only GPT-6 Luna measured
Terminal-Bench-Science (Vals AI)10.0%4.3%GPT-5.6 Terra
Vals Science IndexNot measured28.4%Only GPT-6 Luna measured
τ²-Bench Telecom86.3%Not measuredOnly GPT-5.6 Terra measured
APEX-Agents58.2%44.3%GPT-5.6 Terra
BALROG53.2%Not measuredOnly GPT-5.6 Terra measured
SpireClimbNot measured0.0%Only GPT-6 Luna measured
Vending-Bench 2$7,343Not measuredOnly GPT-5.6 Terra measured
Neither model has a result on the other 14 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index120
Reasoning benchmarks for GPT-5.6 Terra and GPT-6 Luna
BenchmarkGPT-5.6 TerraGPT-6 LunaLeader
GPQA Diamond90.9%Not measuredOnly GPT-5.6 Terra measured
LiveBench reasoning90.6%81.8%GPT-5.6 Terra
MysteryMechanism (Vals AI)Not measured19.4%Only GPT-6 Luna measured
MMMU Pro (Vals AI)86.5%Not measuredOnly GPT-5.6 Terra measured
GPQA Diamond (Artificial Analysis run)92.5%Not measuredOnly GPT-5.6 Terra measured
Humanity's Last Exam (Artificial Analysis run)42.9%38.5%GPT-5.6 Terra
GPQA Diamond (Epoch AI run)93.3%90.5%GPT-5.6 Terra
ARC-AGI96.5%86.7%GPT-5.6 Terra
ARC-AGI-283.9%59.3%GPT-5.6 Terra
SimpleBench48.9%Not measuredOnly GPT-5.6 Terra measured
Chess puzzles47.0%36.0%GPT-5.6 Terra
Mystery game puzzles35.0%7.0%GPT-5.6 Terra
LMArena Text: hard prompts14691410GPT-5.6 Terra
LMArena Text: expert14961442GPT-5.6 Terra
LMArena Text: hard prompts in English14651410GPT-5.6 Terra
LMArena Vision: diagrams12951245GPT-5.6 Terra
LMArena Vision: homework13121261GPT-5.6 Terra
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index60
Math benchmarks for GPT-5.6 Terra and GPT-6 Luna
BenchmarkGPT-5.6 TerraGPT-6 LunaLeader
OTIS Mock AIME99.7%98.9%GPT-5.6 Terra
FrontierMath, tiers 1 to 386.0%78.9%GPT-5.6 Terra
FrontierMath, tier 470.7%56.1%GPT-5.6 Terra
ProofBench74.0%64.0%GPT-5.6 Terra
LMArena Text: math14651409GPT-5.6 Terra
LMArena Text: mathematical work14781427GPT-5.6 Terra
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index51
Knowledge benchmarks for GPT-5.6 Terra and GPT-6 Luna
BenchmarkGPT-5.6 TerraGPT-6 LunaLeader
MMLU Pro86.7%Not measuredOnly GPT-5.6 Terra measured
LegalBench (Vals AI)85.1%Not measuredOnly GPT-5.6 Terra measured
CorpFin (Vals AI)65.3%Not measuredOnly GPT-5.6 Terra measured
TaxEval (Vals AI)76.2%Not measuredOnly GPT-5.6 Terra measured
MortgageTax (Vals AI)67.3%Not measuredOnly GPT-5.6 Terra measured
MedCode (Vals AI)43.4%44.7%GPT-6 Luna
SimpleQA Verified43.2%41.4%GPT-5.6 Terra
LMArena Text: science14521386GPT-5.6 Terra
LMArena Text: medicine and healthcare14301352GPT-5.6 Terra
LMArena Text: legal and government14611400GPT-5.6 Terra
LMArena Text: business and finance14481388GPT-5.6 Terra
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index27
Web design benchmarks for GPT-5.6 Terra and GPT-6 Luna
BenchmarkGPT-5.6 TerraGPT-6 LunaLeader
Design Arena: websites12691272GPT-6 Luna
Design Arena: UI components12591297GPT-6 Luna
Design Arena: game dev12301259GPT-6 Luna
Design Arena: data viz12401215GPT-5.6 Terra
Design Arena: 3D11351217GPT-6 Luna
Design Arena: all code categories12411271GPT-6 Luna
Design Arena: SVG12201253GPT-6 Luna
Design Arena: ASCII art11771156GPT-5.6 Terra
LiveBench design76.0%82.5%GPT-6 Luna
GeneralLike-for-likePublic boards, shown as published, kept out of the index273
General benchmarks for GPT-5.6 Terra and GPT-6 Luna
BenchmarkGPT-5.6 TerraGPT-6 LunaLeader
Vals IndexNot measured52.3%Only GPT-6 Luna measured
LiveBench data analysis79.3%73.4%GPT-5.6 Terra
Public Benefits Bench (Vals AI)62.4%57.6%GPT-5.6 Terra
SAGE (Vals AI)47.0%48.1%GPT-6 Luna
MedScribe (Vals AI)82.9%83.7%GPT-6 Luna
Vals Multimodal Index65.1%Not measuredOnly GPT-5.6 Terra measured
IFBench71.2%Not measuredOnly GPT-5.6 Terra measured
Long context reasoning (AA-LCR)83.0%83.3%GPT-6 Luna
DTBench93.3%90.1%GPT-5.6 Terra
LMCA54.9%44.5%GPT-5.6 Terra
Surface Evolver Bench83.8%Not measuredOnly GPT-5.6 Terra measured
Furniture assembly54.2%44.2%GPT-5.6 Terra
LMArena Text14481389GPT-5.6 Terra
LMArena Text: instruction following14561408GPT-5.6 Terra
LMArena Text: creative writing14131363GPT-5.6 Terra
LMArena Vision12711216GPT-5.6 Terra
LMArena Text: English14491393GPT-5.6 Terra
LMArena Text: non-English14411381GPT-5.6 Terra
LMArena Text: Chinese15141427GPT-5.6 Terra
LMArena Text: German14601376GPT-5.6 Terra
LMArena Text: Japanese14541369GPT-5.6 Terra
LMArena Text: Korean14271359GPT-5.6 Terra
LMArena Text: French14661408GPT-5.6 Terra
LMArena Text: Spanish14511391GPT-5.6 Terra
LMArena Text: Russian14521392GPT-5.6 Terra
LMArena Text: Polish14461383GPT-5.6 Terra
LMArena Text: multi-turn14531392GPT-5.6 Terra
LMArena Text: longer queries14541408GPT-5.6 Terra
LMArena Text: writing and language14311384GPT-5.6 Terra
LMArena Text: entertainment and media14181364GPT-5.6 Terra
LMArena Vision: English12721204GPT-5.6 Terra
LMArena Vision: Chinese13371266GPT-5.6 Terra
LMArena Vision: reading text12781225GPT-5.6 Terra
LMArena Vision: humor1236Not measuredOnly GPT-5.6 Terra measured
LMArena Vision: creative writing12481205GPT-5.6 Terra
LMArena Document1472Not measuredOnly GPT-5.6 Terra measured
Neither model has a result on the other 10 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-5.6 Terra and GPT-6 Luna
BenchmarkGPT-5.6 TerraGPT-6 LunaLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

GPT-5.6 Terra$0.0080Fits in one request
GPT-6 Luna$0.0003Fits in one request

GPT-6 Luna costs less

Repository review

50K input + 3K output tokens

GPT-5.6 Terra$0.136Fits in one request
GPT-6 Luna$0.0065Fits in one request

GPT-6 Luna costs less

Agent loop

200K input + 10K output tokens

GPT-5.6 Terra$0.520Fits in one request
GPT-6 Luna$0.025Fits in one request

GPT-6 Luna costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of GPT-5.6 Terra and GPT-6 Luna
SpecificationOpenAI GPT-5.6 TerraOpenAI GPT-6 Luna
MakerOpenAIOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedJuly 9, 2026September 22, 2026
Context window1.05M1.05M
Input $/Mtok$2.00$0.10
Output $/Mtok$12.00$0.50
Coding index69.071.4
Independent measurements behind it22
Cost per solved task$0.42Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-5.6 Terra better than GPT-6 Luna for coding?

Pick GPT-6 Luna for coding: it scores 71.4 on the Vibe Coding Index against 69.0 for GPT-5.6 Terra. Pick GPT-5.6 Terra when what matters is LiveBench coding or Terminal-Bench 4 (Vals AI), where it leads.

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

GPT-5.6 Terra costs $2.00 per million input tokens and $12.00 per million output tokens; GPT-6 Luna costs $0.10 and $0.50. Cost per solved task is known for GPT-5.6 Terra only.

Which has the bigger context window, GPT-5.6 Terra or GPT-6 Luna?

Neither: both take 1.05M tokens.

How many benchmarks is the GPT-5.6 Terra vs GPT-6 Luna comparison based on?

95 benchmarks have a result for both models as of October 21, 2026. GPT-5.6 Terra's coding index rests on 2 independent benchmarks and GPT-6 Luna's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/gpt-5-6-terra-vs-gpt-6-luna" title="GPT-5.6 Terra vs GPT-6 Luna, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/gpt-5-6-terra-vs-gpt-6-luna">LLMBenchmark.ai, GPT-5.6 Terra vs GPT-6 Luna</a></p>