Enter opens the highlighted result. Esc closes.

Claude Sonnet 5 vs GPT-6 Luna

Claude Sonnet 5 leads GPT-6 Luna on coding, 57.7 to 57.3 on the Vibe Coding Index, and GPT-6 Luna costs 20.0x less per output token ($0.50 vs $10.00 per million). Updated October 22, 2026; 92 benchmarks have a result for both.

Replace Claude Sonnet 5 Replace GPT-6 Luna
Claude Sonnet 5Anthropic
Coding
57.7
Overall
63.2
Output $/M
$10.00
66
GPT-6 LunaOpenAI
Coding
57.3
Overall
59.7
Output $/M
$0.50
25
vs boards won of 92 shared, 1 tied

Claude Sonnet 5 or GPT-6 Luna: which to pick

Claude Sonnet 5 and GPT-6 Luna are level for coding, 57.7 against 57.3 on the Vibe Coding Index, so price and context decide.

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 1M tokens). Pick Claude Sonnet 5 for work measured by Code migration (Vals AI) or Terminal-Bench 2.1 (Vals AI), where it leads.

Model AClaude Sonnet 5Anthropic57.7/1006 independent benchmarks$1.87 per solved taskModel BGPT-6 LunaOpenAI57.3/1005 independent benchmarksNo cost per solved task
Shared results
92
Claude Sonnet 5 only
29
GPT-6 Luna only
11
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
Claude Sonnet 5 57.7 from 6 benchmarks, GPT-6 Luna 57.3 from 5 benchmarks.
Widest category
Math, 12.7 points apart on average over 6 shared benchmarks.
Widest single gap
LiveBench research: Claude Sonnet 5 82.8% against GPT-6 Luna 43.5%, 39.3 points apart on a 0 to 100 scale.
Output price
Claude Sonnet 5 $10.00 against GPT-6 Luna $0.50 per million output tokens, 20.0x apart.
Context window
Claude Sonnet 5 takes 1M 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
CategoryClaude Sonnet 5BarsGPT-6 LunaTies
Coding512
Intelligence20
Agentic work931 tie
Reasoning100
Math33
Knowledge51
Web design63
General263
All shared boards66251 tie
CodingLike-for-likeSix of these boards feed the Vibe Coding Index512
Coding benchmarks for Claude Sonnet 5 and GPT-6 Luna
BenchmarkClaude Sonnet 5GPT-6 LunaLeader
LiveBench coding74.7%84.5%GPT-6 Luna
LMArena WebDev15401578GPT-6 Luna
AA Coding Index71.5Not measuredOnly Claude Sonnet 5 measured
LiveCodeBench (Vals AI)82.4%Not measuredOnly Claude Sonnet 5 measured
SWE-bench Verified (Vals AI)79.6%Not measuredOnly Claude Sonnet 5 measured
Terminal-Bench 4 (Vals AI)10.6%13.6%GPT-6 Luna
Vibe Code Bench (Vals AI)81.3%81.7%GPT-6 Luna
Code migration (Vals AI)44.4%42.5%Claude Sonnet 5
ProgramBench (Vals AI)0.0%0.5%GPT-6 Luna
IOI (Vals AI)45.0%55.6%GPT-6 Luna
Terminal-Bench 2.1 (Vals AI)74.5%73.0%Claude Sonnet 5
SWE-bench Multimodal (Vals AI)Not measured37.5%Only GPT-6 Luna measured
Vibe Code Bench 1-100 (Vals AI)13.8%Not measuredOnly Claude Sonnet 5 measured
SkillsBench (Vals AI)46.5%Not measuredOnly Claude Sonnet 5 measured
CyberBench (Vals AI)70.3%76.3%GPT-6 Luna
SciCode54.3%54.6%GPT-6 Luna
FrontierCode42.7%42.4%Claude Sonnet 5
GSO-Bench37.3%Not measuredOnly Claude Sonnet 5 measured
WeirdML68.8%Not measuredOnly Claude Sonnet 5 measured
ALE-Bench1463.11576.9GPT-6 Luna
LMArena Text: coding14831437Claude Sonnet 5
LMArena Text: software and IT14741423Claude Sonnet 5
LMArena WebDev: React15411573GPT-6 Luna
LMArena WebDev: HTML15271595GPT-6 Luna
LMArena WebDev: from an image15401571GPT-6 Luna
SciCode10.7%Not measuredOnly Claude Sonnet 5 measured
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 Claude Sonnet 5 and GPT-6 Luna
BenchmarkClaude Sonnet 5GPT-6 LunaLeader
AA Intelligence Index38.238.1Claude Sonnet 5
Epoch Capabilities Index156.9156.8Claude Sonnet 5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index93
Agentic work benchmarks for Claude Sonnet 5 and GPT-6 Luna
BenchmarkClaude Sonnet 5GPT-6 LunaLeader
AA Agentic Index43.6Not measuredOnly Claude Sonnet 5 measured
Design Arena: agents building web apps12451191Claude Sonnet 5
Design Arena: agents building full-stack apps12491158Claude Sonnet 5
Design Arena: agents building mobile apps12151134Claude Sonnet 5
Design Arena: agents building Android apps1227Not measuredOnly Claude Sonnet 5 measured
Design Arena: agents building Godot games1225Not measuredOnly Claude Sonnet 5 measured
Design Arena: agents building games1227Not measuredOnly Claude Sonnet 5 measured
Design Arena: HTML slides1168Not measuredOnly Claude Sonnet 5 measured
Design Arena: PowerPoint slides1191Not measuredOnly Claude Sonnet 5 measured
Design Arena: spreadsheets1035Not measuredOnly Claude Sonnet 5 measured
LiveBench research82.8%43.5%Claude Sonnet 5
LiveBench browser72.0%72.0%Tie
LiveBench long agentic39.0%48.1%GPT-6 Luna
SREBench (Vals AI)0.4%2.7%GPT-6 Luna
Finance Agent (Vals AI)53.9%49.9%Claude Sonnet 5
Tax Agent Bench (Vals AI)29.4%19.9%Claude Sonnet 5
Legal Research Bench (Vals AI)41.8%30.3%Claude Sonnet 5
Harvey's Legal Agent Benchmark (Vals AI)Not measured4.2%Only GPT-6 Luna measured
Excel Modeling Benchmark (Vals AI)66.3%68.5%GPT-6 Luna
BioMysteryBench (Vals AI)Not measured61.5%Only GPT-6 Luna measured
Terminal-Bench-Science (Vals AI)5.7%4.3%Claude Sonnet 5
Vals Science IndexNot measured28.4%Only GPT-6 Luna measured
APEX-Agents54.5%44.3%Claude Sonnet 5
SpireClimbNot measured0.0%Only GPT-6 Luna measured
Vending-Bench 2$6,378Not measuredOnly Claude Sonnet 5 measured
Neither model has a result on the other 16 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index100
Reasoning benchmarks for Claude Sonnet 5 and GPT-6 Luna
BenchmarkClaude Sonnet 5GPT-6 LunaLeader
GPQA Diamond88.9%Not measuredOnly Claude Sonnet 5 measured
LiveBench reasoning88.7%81.8%Claude Sonnet 5
MysteryMechanism (Vals AI)Not measured19.4%Only GPT-6 Luna measured
MMMU Pro (Vals AI)83.0%Not measuredOnly Claude Sonnet 5 measured
GPQA Diamond (Artificial Analysis run)91.1%Not measuredOnly Claude Sonnet 5 measured
Humanity's Last Exam (Artificial Analysis run)41.3%38.5%Claude Sonnet 5
GPQA Diamond (Epoch AI run)90.5%90.5%Claude Sonnet 5
ARC-AGINot measured86.7%Only GPT-6 Luna measured
ARC-AGI-2Not measured59.3%Only GPT-6 Luna measured
SimpleBench60.6%Not measuredOnly Claude Sonnet 5 measured
Chess puzzles45.0%36.0%Claude Sonnet 5
Mystery game puzzles35.0%7.0%Claude Sonnet 5
LMArena Text: hard prompts14611410Claude Sonnet 5
LMArena Text: expert14891442Claude Sonnet 5
LMArena Text: hard prompts in English14641410Claude Sonnet 5
LMArena Vision: diagrams12931245Claude Sonnet 5
LMArena Vision: homework13111261Claude Sonnet 5
Neither model has a result on the other 5 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index33
Math benchmarks for Claude Sonnet 5 and GPT-6 Luna
BenchmarkClaude Sonnet 5GPT-6 LunaLeader
OTIS Mock AIME94.7%98.9%GPT-6 Luna
FrontierMath, tiers 1 to 365.6%78.9%GPT-6 Luna
FrontierMath, tier 429.3%56.1%GPT-6 Luna
ProofBench77.0%64.0%Claude Sonnet 5
LMArena Text: math14671409Claude Sonnet 5
LMArena Text: mathematical work14811427Claude Sonnet 5
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 Claude Sonnet 5 and GPT-6 Luna
BenchmarkClaude Sonnet 5GPT-6 LunaLeader
MMLU Pro87.5%Not measuredOnly Claude Sonnet 5 measured
LegalBench (Vals AI)83.9%Not measuredOnly Claude Sonnet 5 measured
CorpFin (Vals AI)67.0%Not measuredOnly Claude Sonnet 5 measured
TaxEval (Vals AI)75.6%Not measuredOnly Claude Sonnet 5 measured
MortgageTax (Vals AI)70.0%Not measuredOnly Claude Sonnet 5 measured
MedCode (Vals AI)47.5%44.7%Claude Sonnet 5
SimpleQA Verified33.7%41.4%GPT-6 Luna
LMArena Text: science14601386Claude Sonnet 5
LMArena Text: medicine and healthcare14421352Claude Sonnet 5
LMArena Text: legal and government14541400Claude Sonnet 5
LMArena Text: business and finance14361388Claude Sonnet 5
FACTS Benchmark SuiteNot measured63.3%Only GPT-6 Luna measured
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index63
Web design benchmarks for Claude Sonnet 5 and GPT-6 Luna
BenchmarkClaude Sonnet 5GPT-6 LunaLeader
Design Arena: websites12741271Claude Sonnet 5
Design Arena: UI components12771295GPT-6 Luna
Design Arena: game dev12771257Claude Sonnet 5
Design Arena: data viz12511216Claude Sonnet 5
Design Arena: 3D12611218Claude Sonnet 5
Design Arena: all code categories12751270Claude Sonnet 5
Design Arena: SVG11921252GPT-6 Luna
Design Arena: ASCII art12071153Claude Sonnet 5
LiveBench design82.0%82.5%GPT-6 Luna
GeneralLike-for-likePublic boards, shown as published, kept out of the index263
General benchmarks for Claude Sonnet 5 and GPT-6 Luna
BenchmarkClaude Sonnet 5GPT-6 LunaLeader
Vals IndexNot measured52.3%Only GPT-6 Luna measured
LiveBench data analysis71.7%73.4%GPT-6 Luna
Public Benefits Bench (Vals AI)66.0%57.6%Claude Sonnet 5
SAGE (Vals AI)48.9%48.1%Claude Sonnet 5
MedScribe (Vals AI)76.0%83.7%GPT-6 Luna
Vals Multimodal Index68.8%Not measuredOnly Claude Sonnet 5 measured
Long context reasoning (AA-LCR)82.0%83.3%GPT-6 Luna
DTBench92.5%90.1%Claude Sonnet 5
LMCA50.0%44.5%Claude Sonnet 5
Surface Evolver Bench60.0%Not measuredOnly Claude Sonnet 5 measured
Furniture assemblyNot measured44.2%Only GPT-6 Luna measured
LMArena Text14421389Claude Sonnet 5
LMArena Text: instruction following14521408Claude Sonnet 5
LMArena Text: creative writing14161363Claude Sonnet 5
LMArena Vision12731216Claude Sonnet 5
LMArena Text: English14501393Claude Sonnet 5
LMArena Text: non-English14311381Claude Sonnet 5
LMArena Text: Chinese14781427Claude Sonnet 5
LMArena Text: German14421376Claude Sonnet 5
LMArena Text: Japanese14221369Claude Sonnet 5
LMArena Text: Korean14121359Claude Sonnet 5
LMArena Text: French14631408Claude Sonnet 5
LMArena Text: Spanish14381391Claude Sonnet 5
LMArena Text: Russian14501392Claude Sonnet 5
LMArena Text: Polish14251383Claude Sonnet 5
LMArena Text: multi-turn14531392Claude Sonnet 5
LMArena Text: longer queries14621408Claude Sonnet 5
LMArena Text: writing and language14361384Claude Sonnet 5
LMArena Text: entertainment and media14111364Claude Sonnet 5
LMArena Vision: English12751204Claude Sonnet 5
LMArena Vision: Chinese13311266Claude Sonnet 5
LMArena Vision: reading text12881225Claude Sonnet 5
LMArena Vision: humor1261Not measuredOnly Claude Sonnet 5 measured
LMArena Vision: creative writing12461205Claude Sonnet 5
LMArena Document1466Not measuredOnly Claude Sonnet 5 measured
Neither model has a result on the other 11 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Sonnet 5 and GPT-6 Luna
BenchmarkClaude Sonnet 5GPT-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

Claude Sonnet 5$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

Claude Sonnet 5$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

Claude Sonnet 5$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 Claude Sonnet 5 and GPT-6 Luna
SpecificationAnthropic Claude Sonnet 5OpenAI GPT-6 Luna
MakerAnthropicOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedJune 30, 2026September 22, 2026
Context window1M1.05M
Input $/Mtok$2.00$0.10
Output $/Mtok$10.00$0.50
Coding index57.757.3
Independent measurements behind it65
Cost per solved task$1.87Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Sonnet 5 better than GPT-6 Luna for coding?

Claude Sonnet 5 and GPT-6 Luna are level for coding, 57.7 against 57.3 on the Vibe Coding Index, so price and context decide. 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 1M tokens). Pick Claude Sonnet 5 for work measured by Code migration (Vals AI) or Terminal-Bench 2.1 (Vals AI), where it leads.

Which is cheaper, Claude Sonnet 5 or GPT-6 Luna?

Claude Sonnet 5 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 Claude Sonnet 5 only.

Which has the bigger context window, Claude Sonnet 5 or GPT-6 Luna?

GPT-6 Luna, with 1.05M tokens against 1M.

How many benchmarks is the Claude Sonnet 5 vs GPT-6 Luna comparison based on?

92 benchmarks have a result for both models as of October 22, 2026. Claude Sonnet 5's coding index rests on 6 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/claude-sonnet-5-vs-gpt-6-luna" title="Claude Sonnet 5 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/claude-sonnet-5-vs-gpt-6-luna">LLMBenchmark.ai, Claude Sonnet 5 vs GPT-6 Luna</a></p>