Enter opens the highlighted result. Esc closes.

Gemma 4 31B vs GPT-6 Luna

GPT-6 Luna leads Gemma 4 31B on coding, 57.3 to 39.0 on the Vibe Coding Index, and Gemma 4 31B costs 1.3x less per output token ($0.40 vs $0.50 per million). Updated October 22, 2026; 55 benchmarks have a result for both.

Replace Gemma 4 31B Replace GPT-6 Luna
Gemma 4 31BGoogle
Coding
39.0
Overall
42.1
Output $/M
$0.40
34
GPT-6 LunaOpenAI
Coding
57.3
Overall
59.7
Output $/M
$0.50
21
vs boards won of 55 shared

Gemma 4 31B or GPT-6 Luna: which to pick

Pick GPT-6 Luna for coding: it scores 57.3 on the Vibe Coding Index against 39.0 for Gemma 4 31B.

Pick Gemma 4 31B for output tokens 1.3x cheaper ($0.40 vs $0.50 per million) and work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.

Model AGemma 4 31BGoogle39.0/1002 independent benchmarksNo cost per solved taskModel BGPT-6 LunaOpenAI57.3/1005 independent benchmarksNo cost per solved task
Shared results
55
Gemma 4 31B only
16
GPT-6 Luna only
48
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
Gemma 4 31B 39.0 from 2 benchmarks, GPT-6 Luna 57.3 from 5 benchmarks.
Widest category
Agentic work, 34.6 points apart on average over 3 shared benchmarks.
Widest single gap
LiveBench browser: Gemma 4 31B 4.0% against GPT-6 Luna 72.0%, 68.0 points apart on a 0 to 100 scale.
Output price
Gemma 4 31B $0.40 against GPT-6 Luna $0.50 per million output tokens, 1.3x apart.
Context window
Gemma 4 31B takes 262K 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
CategoryGemma 4 31BBarsGPT-6 LunaTies
Coding26
Intelligence02
Agentic work12
Reasoning53
Math21
Knowledge42
Web design01
General204
All shared boards3421
CodingLike-for-likeSix of these boards feed the Vibe Coding Index26
Coding benchmarks for Gemma 4 31B and GPT-6 Luna
BenchmarkGemma 4 31BGPT-6 LunaLeader
LiveBench coding49.0%84.5%GPT-6 Luna
LMArena WebDev13671578GPT-6 Luna
AA Coding Index43.4Not measuredOnly Gemma 4 31B measured
Terminal-Bench 4 (Vals AI)Not measured13.6%Only GPT-6 Luna measured
Vibe Code Bench (Vals AI)Not measured81.7%Only GPT-6 Luna measured
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
SciCode45.5%54.6%GPT-6 Luna
Terminal-Bench Hard (Artificial Analysis run)36.4%Not measuredOnly Gemma 4 31B measured
FrontierCodeNot measured42.4%Only GPT-6 Luna measured
WeirdML52.3%Not measuredOnly Gemma 4 31B measured
ALE-Bench925.51576.9GPT-6 Luna
LMArena Text: coding14591437Gemma 4 31B
LMArena Text: software and IT14631423Gemma 4 31B
LMArena WebDev: React13591573GPT-6 Luna
LMArena WebDev: HTML13801595GPT-6 Luna
LMArena WebDev: from an imageNot measured1571Only GPT-6 Luna measured
SciCode8.9%Not measuredOnly Gemma 4 31B measured
Neither model has a result on the other 17 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for Gemma 4 31B and GPT-6 Luna
BenchmarkGemma 4 31BGPT-6 LunaLeader
AA Intelligence Index14.738.1GPT-6 Luna
Epoch Capabilities Index143.0156.8GPT-6 Luna
Agentic workLike-for-likePublic boards, shown as published, kept out of the index12
Agentic work benchmarks for Gemma 4 31B and GPT-6 Luna
BenchmarkGemma 4 31BGPT-6 LunaLeader
AA Agentic Index4.2Not measuredOnly Gemma 4 31B measured
Design Arena: agents building web appsNot measured1191Only GPT-6 Luna measured
Design Arena: agents building full-stack appsNot measured1158Only GPT-6 Luna measured
Design Arena: agents building mobile appsNot measured1134Only GPT-6 Luna measured
LiveBench research44.4%43.5%Gemma 4 31B
LiveBench browser4.0%72.0%GPT-6 Luna
LiveBench long agentic13.3%48.1%GPT-6 Luna
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)Not measured19.9%Only GPT-6 Luna measured
Legal Research Bench (Vals AI)Not measured30.3%Only GPT-6 Luna measured
Harvey's Legal Agent Benchmark (Vals AI)Not measured4.2%Only GPT-6 Luna measured
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
τ²-Bench Telecom65.5%Not measuredOnly Gemma 4 31B measured
APEX-AgentsNot measured44.3%Only GPT-6 Luna measured
SpireClimbNot measured0.0%Only GPT-6 Luna measured
Neither model has a result on the other 22 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index53
Reasoning benchmarks for Gemma 4 31B and GPT-6 Luna
BenchmarkGemma 4 31BGPT-6 LunaLeader
LiveBench reasoning68.3%81.8%GPT-6 Luna
MysteryMechanism (Vals AI)Not measured19.4%Only GPT-6 Luna measured
GPQA Diamond (Artificial Analysis run)85.7%Not measuredOnly Gemma 4 31B measured
Humanity's Last Exam (Artificial Analysis run)23.6%38.5%GPT-6 Luna
GPQA Diamond (Epoch AI run)75.8%90.5%GPT-6 Luna
ARC-AGINot measured86.7%Only GPT-6 Luna measured
ARC-AGI-2Not measured59.3%Only GPT-6 Luna measured
Chess puzzlesNot measured36.0%Only GPT-6 Luna measured
Mystery game puzzlesNot measured7.0%Only GPT-6 Luna measured
LMArena Text: hard prompts14481410Gemma 4 31B
LMArena Text: expert14641442Gemma 4 31B
LMArena Text: hard prompts in English14561410Gemma 4 31B
LMArena Vision: diagrams12831245Gemma 4 31B
LMArena Vision: homework13101261Gemma 4 31B
Kagi LLM Benchmark63.5%Not measuredOnly Gemma 4 31B measured
Neither model has a result on the other 7 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index21
Math benchmarks for Gemma 4 31B and GPT-6 Luna
BenchmarkGemma 4 31BGPT-6 LunaLeader
OTIS Mock AIME73.3%98.9%GPT-6 Luna
FrontierMath, tiers 1 to 3Not measured78.9%Only GPT-6 Luna measured
FrontierMath, tier 4Not measured56.1%Only GPT-6 Luna measured
ProofBenchNot measured64.0%Only GPT-6 Luna measured
LMArena Text: math14631409Gemma 4 31B
LMArena Text: mathematical work14651427Gemma 4 31B
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index42
Knowledge benchmarks for Gemma 4 31B and GPT-6 Luna
BenchmarkGemma 4 31BGPT-6 LunaLeader
CaseLaw (Vals AI)52.6%Not measuredOnly Gemma 4 31B measured
MortgageTax (Vals AI)61.4%Not measuredOnly Gemma 4 31B measured
MedCode (Vals AI)Not measured44.7%Only GPT-6 Luna measured
SimpleQA Verified10.4%41.4%GPT-6 Luna
LMArena Text: science14571386Gemma 4 31B
LMArena Text: medicine and healthcare14511352Gemma 4 31B
LMArena Text: legal and government14381400Gemma 4 31B
LMArena Text: business and finance14301388Gemma 4 31B
FACTS Benchmark Suite45.3%63.3%GPT-6 Luna
Neither model has a result on the other 9 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index01
Web design benchmarks for Gemma 4 31B and GPT-6 Luna
BenchmarkGemma 4 31BGPT-6 LunaLeader
Design Arena: websitesNot measured1271Only GPT-6 Luna measured
Design Arena: UI componentsNot measured1295Only GPT-6 Luna measured
Design Arena: game devNot measured1257Only GPT-6 Luna measured
Design Arena: data vizNot measured1216Only GPT-6 Luna measured
Design Arena: 3DNot measured1218Only GPT-6 Luna measured
Design Arena: all code categoriesNot measured1270Only GPT-6 Luna measured
Design Arena: SVGNot measured1252Only GPT-6 Luna measured
Design Arena: ASCII artNot measured1153Only GPT-6 Luna measured
LiveBench design58.4%82.5%GPT-6 Luna
GeneralLike-for-likePublic boards, shown as published, kept out of the index204
General benchmarks for Gemma 4 31B and GPT-6 Luna
BenchmarkGemma 4 31BGPT-6 LunaLeader
Vals IndexNot measured52.3%Only GPT-6 Luna measured
LiveBench data analysis53.3%73.4%GPT-6 Luna
Public Benefits Bench (Vals AI)Not measured57.6%Only GPT-6 Luna measured
SAGE (Vals AI)55.0%48.1%Gemma 4 31B
MedScribe (Vals AI)Not measured83.7%Only GPT-6 Luna measured
IFBench75.6%Not measuredOnly Gemma 4 31B measured
Long context reasoning (AA-LCR)69.7%83.3%GPT-6 Luna
DTBench82.7%90.1%GPT-6 Luna
LMCA39.3%44.5%GPT-6 Luna
Surface Evolver Bench30.6%Not measuredOnly Gemma 4 31B measured
Furniture assemblyNot measured44.2%Only GPT-6 Luna measured
LMArena Text14431389Gemma 4 31B
LMArena Text: instruction following14331408Gemma 4 31B
LMArena Text: creative writing14171363Gemma 4 31B
LMArena Vision12761216Gemma 4 31B
LMArena Text: English14511393Gemma 4 31B
LMArena Text: non-English14301381Gemma 4 31B
LMArena Text: Chinese14751427Gemma 4 31B
LMArena Text: GermanNot measured1376Only GPT-6 Luna measured
LMArena Text: JapaneseNot measured1369Only GPT-6 Luna measured
LMArena Text: KoreanNot measured1359Only GPT-6 Luna measured
LMArena Text: French14361408Gemma 4 31B
LMArena Text: Spanish14421391Gemma 4 31B
LMArena Text: Russian14601392Gemma 4 31B
LMArena Text: Polish14331383Gemma 4 31B
LMArena Text: multi-turn14521392Gemma 4 31B
LMArena Text: longer queries14461408Gemma 4 31B
LMArena Text: writing and language14321384Gemma 4 31B
LMArena Text: entertainment and media14051364Gemma 4 31B
LMArena Vision: English12711204Gemma 4 31B
LMArena Vision: Chinese13261266Gemma 4 31B
LMArena Vision: reading text12821225Gemma 4 31B
LMArena Vision: humor1280Not measuredOnly Gemma 4 31B measured
LMArena Vision: captioning1218Not measuredOnly Gemma 4 31B measured
LMArena Vision: recognizing things1226Not measuredOnly Gemma 4 31B measured
LMArena Vision: creative writing12851205Gemma 4 31B
LMArena Document1425Not measuredOnly Gemma 4 31B measured
Neither model has a result on the other 9 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemma 4 31B and GPT-6 Luna
BenchmarkGemma 4 31BGPT-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

Gemma 4 31B$0.0003Fits in one request
GPT-6 Luna$0.0003Fits in one request

Gemma 4 31B costs less

Repository review

50K input + 3K output tokens

Gemma 4 31B$0.0082Fits in one request
GPT-6 Luna$0.0065Fits in one request

GPT-6 Luna costs less

Agent loop

200K input + 10K output tokens

Gemma 4 31B$0.032Fits 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 Gemma 4 31B and GPT-6 Luna
SpecificationGoogle Gemma 4 31BOpenAI GPT-6 Luna
MakerGoogleOpenAI
WeightsOpen weightsNo public weights listed
ReleasedApril 2, 2026September 22, 2026
Context window262K1.05M
Input $/Mtok$0.14$0.10
Output $/Mtok$0.40$0.50
Coding index39.057.3
Independent measurements behind it25
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemma 4 31B better than GPT-6 Luna for coding?

Pick GPT-6 Luna for coding: it scores 57.3 on the Vibe Coding Index against 39.0 for Gemma 4 31B. Pick Gemma 4 31B for output tokens 1.3x cheaper ($0.40 vs $0.50 per million) and work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.

Which is cheaper, Gemma 4 31B or GPT-6 Luna?

Gemma 4 31B costs $0.14 per million input tokens and $0.40 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, Gemma 4 31B or GPT-6 Luna?

GPT-6 Luna, with 1.05M tokens against 262K.

How many benchmarks is the Gemma 4 31B vs GPT-6 Luna comparison based on?

55 benchmarks have a result for both models as of October 22, 2026. Gemma 4 31B's coding index rests on 2 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/gemma-4-31b-it-vs-gpt-6-luna" title="Gemma 4 31B 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/gemma-4-31b-it-vs-gpt-6-luna">LLMBenchmark.ai, Gemma 4 31B vs GPT-6 Luna</a></p>