Enter opens the highlighted result. Esc closes.

Gemini 4 Neon vs GPT-6 Luna

Gemini 4 Neon leads GPT-6 Luna on coding, 86.9 to 54.9 on the Vibe Coding Index, and GPT-6 Luna costs 30.0x less per output token ($0.50 vs $15.00 per million). Updated November 22, 2026; 65 benchmarks have a result for both.

Gemini 4 NeonGoogle
Coding
86.9
Overall
89.1
Output $/M
$15.00
65
GPT-6 LunaOpenAI
Coding
54.9
Overall
55.9
Output $/M
$0.50
0
vs boards won of 65 shared

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

Pick Gemini 4 Neon for coding: it scores 86.9 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 30.0x cheaper ($0.50 vs $15.00 per million) and a larger context window (1.05M against 1.05M tokens).

Shared results
65
Gemini 4 Neon only
4
GPT-6 Luna only
52
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 Neon 86.9 from 4 benchmarks, GPT-6 Luna 54.9 from 5 benchmarks.
Widest category
Agentic work, 35.5 points apart on average over 10 shared benchmarks.
Widest single gap
Mystery game puzzles: Gemini 4 Neon 85.0% against GPT-6 Luna 7.0%, 78.0 points apart on a 0 to 100 scale.
Output price
Gemini 4 Neon $15.00 against GPT-6 Luna $0.50 per million output tokens, 30.0x apart.
Context window
Gemini 4 Neon 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 NeonBarsGPT-6 LunaTies
Coding110
Intelligence10
Agentic work100
Reasoning100
Math50
Knowledge60
Web design60
General160
All shared boards650
CodingLike-for-likeSix of these boards feed the Vibe Coding Index110
Coding benchmarks for Gemini 4 Neon and GPT-6 Luna
BenchmarkGemini 4 NeonGPT-6 LunaLeader
LiveBench coding89.7%70.8%Gemini 4 Neon
LMArena WebDev17651578Gemini 4 Neon
Terminal-Bench 4 (Vals AI)61.6%13.6%Gemini 4 Neon
Vibe Code Bench (Vals AI)93.7%81.7%Gemini 4 Neon
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)63.6%12.6%Gemini 4 Neon
FrontierCode 2.044.4%25.2%Gemini 4 Neon
FrontierCodeNot measured42.4%Only GPT-6 Luna measured
MirrorCodeNot measured26.7%Only GPT-6 Luna measured
ALE-BenchNot measured1576.9Only GPT-6 Luna measured
LMArena Text: coding15551434Gemini 4 Neon
LMArena Text: software and IT15421422Gemini 4 Neon
LMArena WebDev: React17871576Gemini 4 Neon
LMArena WebDev: HTML18031593Gemini 4 Neon
LMArena WebDev: from an imageNot measured1567Only GPT-6 Luna measured
GameBench 298.055.8Gemini 4 Neon
Terminal-Bench 4.0Not measured16.4%Only GPT-6 Luna measured
CursorBench 4.065.4%Not measuredOnly Gemini 4 Neon measured
Neither model has a result on the other 28 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for Gemini 4 Neon and GPT-6 Luna
BenchmarkGemini 4 NeonGPT-6 LunaLeader
AA Intelligence Index61.427.9Gemini 4 Neon
Epoch Capabilities IndexNot measured156.4Only GPT-6 Luna measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index100
Agentic work benchmarks for Gemini 4 Neon and GPT-6 Luna
BenchmarkGemini 4 NeonGPT-6 LunaLeader
Design Arena: agents building web appsNot measured1141Only GPT-6 Luna measured
Design Arena: agents building full-stack appsNot measured1140Only GPT-6 Luna measured
Design Arena: agents building mobile appsNot measured1115Only GPT-6 Luna measured
Design Arena: agents building Android appsNot measured1174Only GPT-6 Luna measured
Design Arena: spreadsheetsNot measured871Only GPT-6 Luna measured
LiveBench research94.5%48.5%Gemini 4 Neon
LiveBench browser95.6%70.3%Gemini 4 Neon
LiveBench long agentic82.6%24.7%Gemini 4 Neon
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)54.9%19.9%Gemini 4 Neon
Legal Research Bench (Vals AI)56.3%30.3%Gemini 4 Neon
Harvey's Legal Agent Benchmark (Vals AI)19.6%4.2%Gemini 4 Neon
Excel Modeling Benchmark (Vals AI)Not measured68.5%Only GPT-6 Luna measured
BioMysteryBench (Vals AI)80.0%61.5%Gemini 4 Neon
Terminal-Bench-Science (Vals AI)62.9%4.3%Gemini 4 Neon
Vals Science Index62.6%28.4%Gemini 4 Neon
APEX-Agents82.9%44.3%Gemini 4 Neon
SpireClimbNot measured0.0%Only GPT-6 Luna measured
Vending-Bench 2$16,843Not measuredOnly Gemini 4 Neon measured
LMArena SearchNot measured1157Only 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 index100
Reasoning benchmarks for Gemini 4 Neon and GPT-6 Luna
BenchmarkGemini 4 NeonGPT-6 LunaLeader
LiveBench reasoning89.6%81.8%Gemini 4 Neon
MysteryMechanism (Vals AI)45.0%19.4%Gemini 4 Neon
Humanity's Last Exam (Artificial Analysis run)Not measured38.5%Only GPT-6 Luna measured
HLE-Diamond (Artificial Analysis run)71.3%22.2%Gemini 4 Neon
CritPt v2, hard subset (Artificial Analysis run)61.5%42.5%Gemini 4 Neon
GPQA Diamond (Epoch AI run)96.3%90.5%Gemini 4 Neon
ARC-AGINot measured86.7%Only GPT-6 Luna measured
ARC-AGI-2Not measured59.3%Only GPT-6 Luna measured
SimpleBench94.4%Not measuredOnly Gemini 4 Neon measured
Chess puzzles83.0%36.0%Gemini 4 Neon
Mystery game puzzles85.0%7.0%Gemini 4 Neon
EBR-bench98.1%Not measuredOnly Gemini 4 Neon measured
LMArena Text: hard prompts15471408Gemini 4 Neon
LMArena Text: expert15601437Gemini 4 Neon
LMArena Text: hard prompts in English15241407Gemini 4 Neon
LMArena Vision: diagramsNot measured1238Only GPT-6 Luna measured
LMArena Vision: homeworkNot measured1257Only GPT-6 Luna measured
GoBenchNot measured1119Only 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 index50
Math benchmarks for Gemini 4 Neon and GPT-6 Luna
BenchmarkGemini 4 NeonGPT-6 LunaLeader
OTIS Mock AIME100.0%98.9%Gemini 4 Neon
FrontierMath, tiers 1 to 391.2%78.9%Gemini 4 Neon
FrontierMath, tier 497.6%56.1%Gemini 4 Neon
ProofBenchNot measured64.0%Only GPT-6 Luna measured
LMArena Text: math15481410Gemini 4 Neon
LMArena Text: mathematical work15661424Gemini 4 Neon
Neither model has a result on the other 8 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index60
Knowledge benchmarks for Gemini 4 Neon and GPT-6 Luna
BenchmarkGemini 4 NeonGPT-6 LunaLeader
MedCode (Vals AI)Not measured44.7%Only GPT-6 Luna measured
AA-Omniscience54.8-8.3Gemini 4 Neon
SimpleQA Verified75.2%41.4%Gemini 4 Neon
LMArena Text: science15491387Gemini 4 Neon
LMArena Text: medicine and healthcare15341353Gemini 4 Neon
LMArena Text: legal and government15181393Gemini 4 Neon
LMArena Text: business and finance14931381Gemini 4 Neon
FACTS Benchmark SuiteNot measured64.8%Only GPT-6 Luna measured
Neither model has a result on the other 13 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index60
Web design benchmarks for Gemini 4 Neon and GPT-6 Luna
BenchmarkGemini 4 NeonGPT-6 LunaLeader
Design Arena: websites13471264Gemini 4 Neon
Design Arena: UI components13491283Gemini 4 Neon
Design Arena: game dev13941236Gemini 4 Neon
Design Arena: data vizNot measured1209Only GPT-6 Luna measured
Design Arena: 3D14901182Gemini 4 Neon
Design Arena: all code categories13911254Gemini 4 Neon
Design Arena: SVGNot measured1230Only GPT-6 Luna measured
Design Arena: ASCII artNot measured1150Only GPT-6 Luna measured
LiveBench design81.5%76.1%Gemini 4 Neon
GeneralLike-for-likePublic boards, shown as published, kept out of the index160
General benchmarks for Gemini 4 Neon and GPT-6 Luna
BenchmarkGemini 4 NeonGPT-6 LunaLeader
Vals Index70.9%52.3%Gemini 4 Neon
LiveBench data analysis85.3%73.4%Gemini 4 Neon
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)Not measured83.3%Only GPT-6 Luna measured
GDPval-AANot measured1443Only GPT-6 Luna measured
DTBenchNot measured90.1%Only GPT-6 Luna measured
LMCANot measured44.5%Only GPT-6 Luna measured
Furniture assembly89.2%44.2%Gemini 4 Neon
LMArena Text15241390Gemini 4 Neon
LMArena Text: instruction following15351405Gemini 4 Neon
LMArena Text: creative writing15191360Gemini 4 Neon
LMArena VisionNot measured1210Only GPT-6 Luna measured
LMArena Text: English15111392Gemini 4 Neon
LMArena Text: non-English15241381Gemini 4 Neon
LMArena Text: Chinese15981423Gemini 4 Neon
LMArena Text: GermanNot measured1381Only GPT-6 Luna measured
LMArena Text: JapaneseNot measured1363Only GPT-6 Luna measured
LMArena Text: KoreanNot measured1349Only GPT-6 Luna measured
LMArena Text: French15631403Gemini 4 Neon
LMArena Text: Spanish15051396Gemini 4 Neon
LMArena Text: Russian15541390Gemini 4 Neon
LMArena Text: PolishNot measured1381Only GPT-6 Luna measured
LMArena Text: multi-turn15251395Gemini 4 Neon
LMArena Text: longer queries15331404Gemini 4 Neon
LMArena Text: writing and language15171383Gemini 4 Neon
LMArena Text: entertainment and media14951359Gemini 4 Neon
LMArena Vision: EnglishNot measured1208Only GPT-6 Luna measured
LMArena Vision: ChineseNot measured1273Only GPT-6 Luna measured
LMArena Vision: reading textNot measured1220Only GPT-6 Luna measured
LMArena Vision: humorNot measured1167Only GPT-6 Luna measured
LMArena Vision: creative writingNot measured1198Only 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 Neon and GPT-6 Luna
BenchmarkGemini 4 NeonGPT-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 Neon$0.011Fits 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 Neon$0.195Fits 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 Neon$0.750Fits 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 Neon and GPT-6 Luna
SpecificationGoogle Gemini 4 NeonOpenAI GPT-6 Luna
MakerGoogleOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedNovember 18, 2026September 22, 2026
Context window1.05M1.05M
Input $/Mtok$3.00$0.10
Output $/Mtok$15.00$0.50
Coding index86.954.9
Independent measurements behind it45
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

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

Pick Gemini 4 Neon for coding: it scores 86.9 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 30.0x cheaper ($0.50 vs $15.00 per million) and a larger context window (1.05M against 1.05M tokens).

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

Gemini 4 Neon costs $3.00 per million input tokens and $15.00 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 Neon or GPT-6 Luna?

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

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

65 benchmarks have a result for both models as of November 22, 2026. Gemini 4 Neon'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-neon-vs-gpt-6-luna" title="Gemini 4 Neon 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-neon-vs-gpt-6-luna">LLMBenchmark.ai, Gemini 4 Neon vs GPT-6 Luna</a></p>