Enter opens the highlighted result. Esc closes.

GPT-6 Luna vs GPT-6 Sol

GPT-6 Sol leads GPT-6 Luna on coding, 76.0 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; 99 benchmarks have a result for both.

Replace GPT-6 Luna Replace GPT-6 Sol
GPT-6 LunaOpenAI
Coding
57.3
Overall
59.7
Output $/M
$0.50
18
GPT-6 SolOpenAI
Coding
76.0
Overall
74.6
Output $/M
$10.00
81
vs boards won of 99 shared

GPT-6 Luna or GPT-6 Sol: which to pick

Pick GPT-6 Sol for coding: it scores 76.0 on the Vibe Coding Index against 57.3 for GPT-6 Luna.

Pick GPT-6 Luna for output tokens 20.0x cheaper ($0.50 vs $10.00 per million) and work measured by Finance Agent (Vals AI) or Tax Agent Bench (Vals AI), where it leads.

Model AGPT-6 LunaOpenAI57.3/1005 independent benchmarksNo cost per solved taskModel BGPT-6 SolOpenAI76.0/1005 independent benchmarksNo cost per solved task
Shared results
99
GPT-6 Luna only
4
GPT-6 Sol only
2
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
GPT-6 Luna 57.3 from 5 benchmarks, GPT-6 Sol 76.0 from 5 benchmarks.
Widest category
Agentic work, 12.6 points apart on average over 16 shared benchmarks.
Widest single gap
Mystery game puzzles: GPT-6 Luna 7.0% against GPT-6 Sol 56.0%, 49.0 points apart on a 0 to 100 scale.
Output price
GPT-6 Luna $0.50 against GPT-6 Sol $10.00 per million output tokens, 20.0x 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-6 LunaBarsGPT-6 SolTies
Coding017
Intelligence02
Agentic work313
Reasoning112
Math24
Knowledge25
Web design27
General821
All shared boards1881
CodingLike-for-likeSix of these boards feed the Vibe Coding Index017
Coding benchmarks for GPT-6 Luna and GPT-6 Sol
BenchmarkGPT-6 LunaGPT-6 SolLeader
LiveBench coding84.5%93.1%GPT-6 Sol
LMArena WebDev15781683GPT-6 Sol
Terminal-Bench 4 (Vals AI)13.6%44.4%GPT-6 Sol
Vibe Code Bench (Vals AI)81.7%87.8%GPT-6 Sol
Code migration (Vals AI)42.5%57.2%GPT-6 Sol
ProgramBench (Vals AI)0.5%2.0%GPT-6 Sol
IOI (Vals AI)55.6%82.6%GPT-6 Sol
Terminal-Bench 2.1 (Vals AI)73.0%83.2%GPT-6 Sol
SWE-bench Multimodal (Vals AI)37.5%Not measuredOnly GPT-6 Luna measured
CyberBench (Vals AI)76.3%78.0%GPT-6 Sol
SciCode54.6%57.6%GPT-6 Sol
FrontierCode42.4%49.3%GPT-6 Sol
ALE-Bench1576.92462.0GPT-6 Sol
LMArena Text: coding14371448GPT-6 Sol
LMArena Text: software and IT14231433GPT-6 Sol
LMArena WebDev: React15731685GPT-6 Sol
LMArena WebDev: HTML15951685GPT-6 Sol
LMArena WebDev: from an image15711660GPT-6 Sol
Neither model has a result on the other 21 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for GPT-6 Luna and GPT-6 Sol
BenchmarkGPT-6 LunaGPT-6 SolLeader
AA Intelligence Index38.147.6GPT-6 Sol
Epoch Capabilities Index156.8163.7GPT-6 Sol
Agentic workLike-for-likePublic boards, shown as published, kept out of the index313
Agentic work benchmarks for GPT-6 Luna and GPT-6 Sol
BenchmarkGPT-6 LunaGPT-6 SolLeader
Design Arena: agents building web apps11911231GPT-6 Sol
Design Arena: agents building full-stack apps11581238GPT-6 Sol
Design Arena: agents building mobile apps11341217GPT-6 Sol
LiveBench research43.5%69.3%GPT-6 Sol
LiveBench browser72.0%92.0%GPT-6 Sol
LiveBench long agentic48.1%66.9%GPT-6 Sol
SREBench (Vals AI)2.7%30.5%GPT-6 Sol
Finance Agent (Vals AI)49.9%49.0%GPT-6 Luna
Tax Agent Bench (Vals AI)19.9%15.0%GPT-6 Luna
Legal Research Bench (Vals AI)30.3%28.9%GPT-6 Luna
Harvey's Legal Agent Benchmark (Vals AI)4.2%Not measuredOnly GPT-6 Luna measured
Excel Modeling Benchmark (Vals AI)68.5%71.5%GPT-6 Sol
BioMysteryBench (Vals AI)61.5%74.8%GPT-6 Sol
Terminal-Bench-Science (Vals AI)4.3%30.0%GPT-6 Sol
Vals Science Index28.4%45.0%GPT-6 Sol
APEX-Agents44.3%54.3%GPT-6 Sol
SpireClimb0.0%7.5%GPT-6 Sol
Vending-Bench 2Not measured$14,428Only GPT-6 Sol measured
Neither model has a result on the other 23 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index112
Reasoning benchmarks for GPT-6 Luna and GPT-6 Sol
BenchmarkGPT-6 LunaGPT-6 SolLeader
LiveBench reasoning81.8%88.7%GPT-6 Sol
MysteryMechanism (Vals AI)19.4%30.2%GPT-6 Sol
Humanity's Last Exam (Artificial Analysis run)38.5%47.9%GPT-6 Sol
GPQA Diamond (Epoch AI run)90.5%94.3%GPT-6 Sol
ARC-AGI86.7%95.5%GPT-6 Sol
ARC-AGI-259.3%89.6%GPT-6 Sol
Chess puzzles36.0%56.0%GPT-6 Sol
Mystery game puzzles7.0%56.0%GPT-6 Sol
EBR-benchNot measured53.3%Only GPT-6 Sol measured
LMArena Text: hard prompts14101422GPT-6 Sol
LMArena Text: expert14421437GPT-6 Luna
LMArena Text: hard prompts in English14101420GPT-6 Sol
LMArena Vision: diagrams12451270GPT-6 Sol
LMArena Vision: homework12611286GPT-6 Sol
Neither model has a result on the other 8 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index24
Math benchmarks for GPT-6 Luna and GPT-6 Sol
BenchmarkGPT-6 LunaGPT-6 SolLeader
OTIS Mock AIME98.9%100.0%GPT-6 Sol
FrontierMath, tiers 1 to 378.9%89.8%GPT-6 Sol
FrontierMath, tier 456.1%90.0%GPT-6 Sol
ProofBench64.0%83.0%GPT-6 Sol
LMArena Text: math14091407GPT-6 Luna
LMArena Text: mathematical work14271424GPT-6 Luna
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index25
Knowledge benchmarks for GPT-6 Luna and GPT-6 Sol
BenchmarkGPT-6 LunaGPT-6 SolLeader
MedCode (Vals AI)44.7%47.1%GPT-6 Sol
SimpleQA Verified41.4%60.7%GPT-6 Sol
LMArena Text: science13861389GPT-6 Sol
LMArena Text: medicine and healthcare13521366GPT-6 Sol
LMArena Text: legal and government14001396GPT-6 Luna
LMArena Text: business and finance13881380GPT-6 Luna
FACTS Benchmark Suite63.3%72.5%GPT-6 Sol
Neither model has a result on the other 11 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index27
Web design benchmarks for GPT-6 Luna and GPT-6 Sol
BenchmarkGPT-6 LunaGPT-6 SolLeader
Design Arena: websites12711292GPT-6 Sol
Design Arena: UI components12951290GPT-6 Luna
Design Arena: game dev12571311GPT-6 Sol
Design Arena: data viz12161269GPT-6 Sol
Design Arena: 3D12181314GPT-6 Sol
Design Arena: all code categories12701300GPT-6 Sol
Design Arena: SVG12521315GPT-6 Sol
Design Arena: ASCII art11531222GPT-6 Sol
LiveBench design82.5%82.2%GPT-6 Luna
GeneralLike-for-likePublic boards, shown as published, kept out of the index821
General benchmarks for GPT-6 Luna and GPT-6 Sol
BenchmarkGPT-6 LunaGPT-6 SolLeader
Vals Index52.3%Not measuredOnly GPT-6 Luna measured
LiveBench data analysis73.4%81.2%GPT-6 Sol
Public Benefits Bench (Vals AI)57.6%56.6%GPT-6 Luna
SAGE (Vals AI)48.1%44.8%GPT-6 Luna
MedScribe (Vals AI)83.7%82.0%GPT-6 Luna
Long context reasoning (AA-LCR)83.3%83.7%GPT-6 Sol
DTBench90.1%97.3%GPT-6 Sol
LMCA44.5%59.1%GPT-6 Sol
Furniture assembly44.2%58.3%GPT-6 Sol
LMArena Text13891398GPT-6 Sol
LMArena Text: instruction following14081413GPT-6 Sol
LMArena Text: creative writing13631378GPT-6 Sol
LMArena Vision12161240GPT-6 Sol
LMArena Text: English13931396GPT-6 Sol
LMArena Text: non-English13811388GPT-6 Sol
LMArena Text: Chinese14271412GPT-6 Luna
LMArena Text: German13761379GPT-6 Sol
LMArena Text: Japanese13691377GPT-6 Sol
LMArena Text: Korean13591336GPT-6 Luna
LMArena Text: French14081399GPT-6 Luna
LMArena Text: Spanish13911387GPT-6 Luna
LMArena Text: Russian13921402GPT-6 Sol
LMArena Text: Polish13831372GPT-6 Luna
LMArena Text: multi-turn13921414GPT-6 Sol
LMArena Text: longer queries14081416GPT-6 Sol
LMArena Text: writing and language13841389GPT-6 Sol
LMArena Text: entertainment and media13641373GPT-6 Sol
LMArena Vision: English12041229GPT-6 Sol
LMArena Vision: Chinese1266Not measuredOnly GPT-6 Luna measured
LMArena Vision: reading text12251258GPT-6 Sol
LMArena Vision: creative writing12051219GPT-6 Sol
Neither model has a result on the other 15 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-6 Luna and GPT-6 Sol
BenchmarkGPT-6 LunaGPT-6 SolLeader
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-6 Luna$0.0003Fits in one request
GPT-6 Sol$0.0070Fits in one request

GPT-6 Luna costs less

Repository review

50K input + 3K output tokens

GPT-6 Luna$0.0065Fits in one request
GPT-6 Sol$0.130Fits in one request

GPT-6 Luna costs less

Agent loop

200K input + 10K output tokens

GPT-6 Luna$0.025Fits in one request
GPT-6 Sol$0.500Fits 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-6 Luna and GPT-6 Sol
SpecificationOpenAI GPT-6 LunaOpenAI GPT-6 Sol
MakerOpenAIOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedSeptember 22, 2026September 22, 2026
Context window1.05M1.05M
Input $/Mtok$0.10$2.00
Output $/Mtok$0.50$10.00
Coding index57.376.0
Independent measurements behind it55
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-6 Luna better than GPT-6 Sol for coding?

Pick GPT-6 Sol for coding: it scores 76.0 on the Vibe Coding Index against 57.3 for GPT-6 Luna. Pick GPT-6 Luna for output tokens 20.0x cheaper ($0.50 vs $10.00 per million) and work measured by Finance Agent (Vals AI) or Tax Agent Bench (Vals AI), where it leads.

Which is cheaper, GPT-6 Luna or GPT-6 Sol?

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

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

Neither: both take 1.05M tokens.

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

99 benchmarks have a result for both models as of October 22, 2026. GPT-6 Luna's coding index rests on 5 independent benchmarks and GPT-6 Sol'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/gpt-6-luna-vs-gpt-6-sol" title="GPT-6 Luna vs GPT-6 Sol, 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-6-luna-vs-gpt-6-sol">LLMBenchmark.ai, GPT-6 Luna vs GPT-6 Sol</a></p>