Enter opens the highlighted result. Esc closes.

Gemini 4 Carbon vs GPT-6 Luna

Gemini 4 Carbon leads GPT-6 Luna on coding, 86.6 to 54.9 on the Vibe Coding Index, and GPT-6 Luna costs 20.0x less per output token ($0.50 vs $10.00 per million). Updated November 14, 2026; 67 benchmarks have a result for both.

Replace Gemini 4 Carbon Replace GPT-6 Luna
Gemini 4 CarbonGoogle
Coding
86.6
Overall
88.3
Output $/M
$10.00
66
GPT-6 LunaOpenAI
Coding
54.9
Overall
55.9
Output $/M
$0.50
1
vs boards won of 67 shared

Gemini 4 Carbon or GPT-6 Luna: which to pick

Pick Gemini 4 Carbon for coding: it scores 86.6 on the Vibe Coding Index against 54.9 for GPT-6 Luna, though on 4 benchmarks against 5.

Pick GPT-6 Luna for output tokens 20.0x cheaper ($0.50 vs $10.00 per million) and a larger context window (1.05M against 1.05M tokens).

Model AGemini 4 CarbonGoogle86.6/1004 independent benchmarksNo cost per solved taskModel BGPT-6 LunaOpenAI54.9/1005 independent benchmarksNo cost per solved task
Shared results
67
Gemini 4 Carbon only
6
GPT-6 Luna only
51
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

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
Gemini 4 Carbon 86.6 from 4 benchmarks, GPT-6 Luna 54.9 from 5 benchmarks.
Widest category
Agentic work, 30.7 points apart on average over 9 shared benchmarks.
Widest single gap
Mystery game puzzles: Gemini 4 Carbon 73.0% against GPT-6 Luna 7.0%, 66.0 points apart on a 0 to 100 scale.
Output price
Gemini 4 Carbon $10.00 against GPT-6 Luna $0.50 per million output tokens, 20.0x apart.
Context window
Gemini 4 Carbon takes 1.05M tokens, GPT-6 Luna 1.05M.

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
CategoryGemini 4 CarbonBarsGPT-6 LunaTies
Coding110
Intelligence20
Agentic work90
Reasoning110
Math40
Knowledge80
Web design60
General151
All shared boards661
CodingLike-for-likeSix of these boards feed the Vibe Coding Index110
Coding benchmarks for Gemini 4 Carbon and GPT-6 Luna
BenchmarkGemini 4 CarbonGPT-6 LunaLeader
LiveBench coding91.0%70.8%Gemini 4 Carbon
LMArena WebDev17681577Gemini 4 Carbon
Terminal-Bench 4 (Vals AI)63.1%13.6%Gemini 4 Carbon
Vibe Code Bench (Vals AI)89.1%81.7%Gemini 4 Carbon
Code migration (Vals AI)Not measured42.5%Only GPT-6 Luna measured
ProgramBench (Vals AI)Not measured0.5%Only GPT-6 Luna measured
IOI (Vals AI)Not measured55.6%Only GPT-6 Luna measured
Terminal-Bench 2.1 (Vals AI)Not measured73.0%Only GPT-6 Luna measured
SWE-bench Multimodal (Vals AI)Not measured37.5%Only GPT-6 Luna measured
CyberBench (Vals AI)Not measured76.3%Only GPT-6 Luna measured
SciCodeNot measured54.6%Only GPT-6 Luna measured
Terminal-Bench 4.0 (Artificial Analysis run)65.2%12.6%Gemini 4 Carbon
FrontierCode 2.040.6%25.2%Gemini 4 Carbon
FrontierCodeNot measured42.4%Only GPT-6 Luna measured
FrontierSWE61.2%Not measuredOnly Gemini 4 Carbon measured
MirrorCodeNot measured26.7%Only GPT-6 Luna measured
ALE-Bench2374.71576.9Gemini 4 Carbon
LMArena Text: coding15641431Gemini 4 Carbon
LMArena Text: software and IT15521422Gemini 4 Carbon
LMArena WebDev: React17921574Gemini 4 Carbon
LMArena WebDev: HTML17701595Gemini 4 Carbon
LMArena WebDev: from an imageNot measured1566Only GPT-6 Luna measured
GameBench 2Not measured55.9Only GPT-6 Luna measured
Terminal-Bench 4.0Not measured16.4%Only GPT-6 Luna measured
CursorBench 4.060.4%Not measuredOnly Gemini 4 Carbon measured
Neither model has a result on the other 27 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for Gemini 4 Carbon and GPT-6 Luna
BenchmarkGemini 4 CarbonGPT-6 LunaLeader
AA Intelligence Index60.027.9Gemini 4 Carbon
Epoch Capabilities Index169.8156.4Gemini 4 Carbon
Agentic workLike-for-likePublic boards, shown as published, kept out of the index90
Agentic work benchmarks for Gemini 4 Carbon and GPT-6 Luna
BenchmarkGemini 4 CarbonGPT-6 LunaLeader
Design Arena: agents building web appsNot measured1150Only GPT-6 Luna measured
Design Arena: agents building full-stack appsNot measured1138Only GPT-6 Luna measured
Design Arena: agents building mobile apps12841117Gemini 4 Carbon
Design Arena: agents building Android appsNot measured1176Only GPT-6 Luna measured
Design Arena: spreadsheets1224875Gemini 4 Carbon
LiveBench research92.5%48.5%Gemini 4 Carbon
LiveBench browser80.0%70.3%Gemini 4 Carbon
LiveBench long agentic72.7%24.7%Gemini 4 Carbon
SREBench (Vals AI)Not measured2.7%Only GPT-6 Luna measured
Finance Agent (Vals AI)Not measured49.9%Only GPT-6 Luna measured
Tax Agent Bench (Vals AI)51.8%19.9%Gemini 4 Carbon
Legal Research Bench (Vals AI)55.8%30.3%Gemini 4 Carbon
Harvey's Legal Agent Benchmark (Vals AI)17.5%4.2%Gemini 4 Carbon
Excel Modeling Benchmark (Vals AI)Not measured68.5%Only GPT-6 Luna measured
BioMysteryBench (Vals AI)Not measured61.5%Only GPT-6 Luna measured
Terminal-Bench-Science (Vals AI)Not measured4.3%Only GPT-6 Luna measured
Vals Science IndexNot measured28.4%Only GPT-6 Luna measured
APEX-Agents84.2%44.3%Gemini 4 Carbon
SpireClimbNot measured0.0%Only GPT-6 Luna measured
Vending-Bench 2$16,484Not measuredOnly Gemini 4 Carbon measured
LMArena SearchNot measured1160Only GPT-6 Luna measured
Neither model has a result on the other 23 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index110
Reasoning benchmarks for Gemini 4 Carbon and GPT-6 Luna
BenchmarkGemini 4 CarbonGPT-6 LunaLeader
LiveBench reasoning90.6%81.8%Gemini 4 Carbon
MysteryMechanism (Vals AI)Not measured19.4%Only GPT-6 Luna measured
Humanity's Last Exam (Artificial Analysis run)Not measured38.5%Only GPT-6 Luna measured
HLE-Diamond (Artificial Analysis run)69.1%22.2%Gemini 4 Carbon
CritPt v2, hard subset (Artificial Analysis run)69.0%42.5%Gemini 4 Carbon
GPQA Diamond (Epoch AI run)96.5%90.5%Gemini 4 Carbon
ARC-AGI98.5%86.7%Gemini 4 Carbon
ARC-AGI-294.2%59.3%Gemini 4 Carbon
SimpleBench92.6%Not measuredOnly Gemini 4 Carbon measured
Chess puzzles78.0%36.0%Gemini 4 Carbon
Mystery game puzzles73.0%7.0%Gemini 4 Carbon
EBR-bench87.6%Not measuredOnly Gemini 4 Carbon measured
LMArena Text: hard prompts15511407Gemini 4 Carbon
LMArena Text: expert15821437Gemini 4 Carbon
LMArena Text: hard prompts in English15451407Gemini 4 Carbon
LMArena Vision: diagramsNot measured1235Only GPT-6 Luna measured
LMArena Vision: homeworkNot measured1252Only GPT-6 Luna measured
GoBenchNot measured1120Only GPT-6 Luna measured
Neither model has a result on the other 8 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index40
Math benchmarks for Gemini 4 Carbon and GPT-6 Luna
BenchmarkGemini 4 CarbonGPT-6 LunaLeader
OTIS Mock AIME100.0%98.9%Gemini 4 Carbon
FrontierMath, tiers 1 to 390.5%78.9%Gemini 4 Carbon
FrontierMath, tier 495.1%56.1%Gemini 4 Carbon
ProofBenchNot measured64.0%Only GPT-6 Luna measured
LMArena Text: mathNot measured1411Only GPT-6 Luna measured
LMArena Text: mathematical work15381425Gemini 4 Carbon
MathArena85.4%Not measuredOnly Gemini 4 Carbon measured
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index80
Knowledge benchmarks for Gemini 4 Carbon and GPT-6 Luna
BenchmarkGemini 4 CarbonGPT-6 LunaLeader
MedCode (Vals AI)Not measured44.7%Only GPT-6 Luna measured
AA-Omniscience53.5-8.3Gemini 4 Carbon
SimpleQA Verified77.7%41.4%Gemini 4 Carbon
LMArena Text: science15351387Gemini 4 Carbon
LMArena Text: medicine and healthcare15251349Gemini 4 Carbon
LMArena Text: legal and government15421392Gemini 4 Carbon
LMArena Text: business and finance15201382Gemini 4 Carbon
FACTS Benchmark Suite78.4%64.8%Gemini 4 Carbon
PRBench Legal53.3%42.9%Gemini 4 Carbon
Neither model has a result on the other 12 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index60
Web design benchmarks for Gemini 4 Carbon and GPT-6 Luna
BenchmarkGemini 4 CarbonGPT-6 LunaLeader
Design Arena: websites13191268Gemini 4 Carbon
Design Arena: UI components13571291Gemini 4 Carbon
Design Arena: game dev13261253Gemini 4 Carbon
Design Arena: data vizNot measured1208Only GPT-6 Luna measured
Design Arena: 3D14341194Gemini 4 Carbon
Design Arena: all code categories13481261Gemini 4 Carbon
Design Arena: SVGNot measured1235Only GPT-6 Luna measured
Design Arena: ASCII artNot measured1161Only GPT-6 Luna measured
LiveBench design80.2%76.1%Gemini 4 Carbon
GeneralLike-for-likePublic boards, shown as published, kept out of the index151
General benchmarks for Gemini 4 Carbon and GPT-6 Luna
BenchmarkGemini 4 CarbonGPT-6 LunaLeader
Vals Index70.5%52.3%Gemini 4 Carbon
LiveBench data analysis86.5%73.4%Gemini 4 Carbon
Public Benefits Bench (Vals AI)Not measured57.6%Only GPT-6 Luna measured
SAGE (Vals AI)Not measured83.4%Only GPT-6 Luna measured
MedScribe (Vals AI)Not measured83.7%Only GPT-6 Luna measured
Long context reasoning (AA-LCR)79.3%83.3%GPT-6 Luna
GDPval-AA17331443Gemini 4 Carbon
DTBenchNot measured90.1%Only GPT-6 Luna measured
LMCANot measured44.5%Only GPT-6 Luna measured
Furniture assembly77.5%44.2%Gemini 4 Carbon
LMArena Text15381388Gemini 4 Carbon
LMArena Text: instruction following15391403Gemini 4 Carbon
LMArena Text: creative writing15331359Gemini 4 Carbon
LMArena VisionNot measured1211Only GPT-6 Luna measured
LMArena Text: English15301391Gemini 4 Carbon
LMArena Text: non-English15341380Gemini 4 Carbon
LMArena Text: Chinese16131422Gemini 4 Carbon
LMArena Text: GermanNot measured1384Only GPT-6 Luna measured
LMArena Text: JapaneseNot measured1366Only GPT-6 Luna measured
LMArena Text: KoreanNot measured1347Only GPT-6 Luna measured
LMArena Text: FrenchNot measured1405Only GPT-6 Luna measured
LMArena Text: SpanishNot measured1395Only GPT-6 Luna measured
LMArena Text: Russian15601388Gemini 4 Carbon
LMArena Text: PolishNot measured1382Only GPT-6 Luna measured
LMArena Text: multi-turn15691393Gemini 4 Carbon
LMArena Text: longer queries15491403Gemini 4 Carbon
LMArena Text: writing and language15421382Gemini 4 Carbon
LMArena Text: entertainment and media15271359Gemini 4 Carbon
LMArena Vision: EnglishNot measured1204Only GPT-6 Luna measured
LMArena Vision: ChineseNot measured1274Only GPT-6 Luna measured
LMArena Vision: reading textNot measured1221Only GPT-6 Luna measured
LMArena Vision: humorNot measured1168Only GPT-6 Luna measured
LMArena Vision: creative writingNot measured1190Only GPT-6 Luna measured
Neither model has a result on the other 16 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemini 4 Carbon and GPT-6 Luna
BenchmarkGemini 4 CarbonGPT-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

Gemini 4 Carbon$0.0070Fits in one request
GPT-6 Luna$0.0003Fits in one request

GPT-6 Luna costs less

Repository review

50K input + 3K output tokens

Gemini 4 Carbon$0.130Fits in one request
GPT-6 Luna$0.0065Fits in one request

GPT-6 Luna costs less

Agent loop

200K input + 10K output tokens

Gemini 4 Carbon$0.500Fits 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 Gemini 4 Carbon and GPT-6 Luna
SpecificationGoogle Gemini 4 CarbonOpenAI GPT-6 Luna
MakerGoogleOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedNovember 10, 2026September 22, 2026
Context window1.05M1.05M
Input $/Mtok$2.00$0.10
Output $/Mtok$10.00$0.50
Coding index86.654.9
Independent measurements behind it45
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemini 4 Carbon better than GPT-6 Luna for coding?

Pick Gemini 4 Carbon for coding: it scores 86.6 on the Vibe Coding Index against 54.9 for GPT-6 Luna, though on 4 benchmarks against 5. Pick GPT-6 Luna for output tokens 20.0x cheaper ($0.50 vs $10.00 per million) and a larger context window (1.05M against 1.05M tokens).

Which is cheaper, Gemini 4 Carbon or GPT-6 Luna?

Gemini 4 Carbon costs $2.00 per million input tokens and $10.00 per million output tokens; GPT-6 Luna costs $0.10 and $0.50. Cost per solved task is known for neither of them.

Which has the bigger context window, Gemini 4 Carbon or GPT-6 Luna?

GPT-6 Luna, with 1.05M tokens against 1.05M.

How many benchmarks is the Gemini 4 Carbon vs GPT-6 Luna comparison based on?

67 benchmarks have a result for both models as of November 14, 2026. Gemini 4 Carbon's coding index rests on 4 independent benchmarks and GPT-6 Luna's on 5. 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/gemini-4-carbon-vs-gpt-6-luna" title="Gemini 4 Carbon 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/gemini-4-carbon-vs-gpt-6-luna">LLMBenchmark.ai, Gemini 4 Carbon vs GPT-6 Luna</a></p>