OpenAILLM matchup

GPT-6 Solvs.GPT-6 Luna

Is the higher price worth it?

GPT-6 Sol performs better than GPT-6 Luna, but costs 20 times as much.

GPT-6 Sol ahead

IntelligenceOverall index, 100 points = average

110.5

GPT-6 Sol

Rank 13

96.3

GPT-6 Luna

Rank 35

about as fast

SpeedTokens per second

111

GPT-6 Sol

101

GPT-6 Luna

GPT-6 Luna cheaper

Price$ per 1M tokens

GPT-6 Sol

$0.20

GPT-6 Luna

GPT-6 Sol more popular

Popularity0 to 100

GPT-6 Sol

Rank 3 of 785

GPT-6 Luna

Rank 30 of 785

Above 272,000 tokens of context, GPT-6 Sol costs more: $4 input and $15 output per 1M tokens. Above 272,000 tokens of context, GPT-6 Luna costs more: $0.20 input and $0.75 output per 1M tokens.

Choose GPT-6 Sol for

KnowledgeGeneralCodingReasoningSearchMultimodalCybersecurityLegalMathematicsHealthProductivityAgentsFinance (not certain)

Choose GPT-6 Luna for

No clear advantage

Where is the difference?

Strengths at a glance

GPT-6 SolGPT-6 LunaMedian of all ranked models

CodingGPT-6 Sol ahead

GPT-6 Sol114.9Rank 11

GPT-6 Luna97.5Rank 35

Median of ranked models84.1

95 % range: GPT-6 Sol 112.2 to 117.6, GPT-6 Luna 94.5 to 100.5.

AreaGPT-6 SolGPT-6 Luna

The same values and verdicts as the bars below. In the close-up the net starts at 80 points, just below the weaker model's lowest value; the reference group average is 100.

GPT-6 Sol is ahead in 9 of 10 areas, GPT-6 Luna in none.

GPT-6 Sol ahead: 9GPT-6 Sol slightly ahead: 1

Areas in the overall index

KnowledgeGPT-6 Sol +21109.8 to 89.2

GeneralGPT-6 Sol +19113.0 to 94.4

CodingGPT-6 Sol +17114.9 to 97.5

ReasoningGPT-6 Sol +16114.1 to 97.8

MultimodalGPT-6 Sol +15106.9 to 92.0

MathematicsGPT-6 Sol +12117.5 to 105.5

AgentsGPT-6 Sol +8107.2 to 99.5

Specialist fields

LegalGPT-6 Sol +14104.8 to 90.7

HealthGPT-6 Sol +10104.8 to 95.2

Financeleaning GPT-6 Solnot certain · 3 and 5 tests

Bar length shows the gap in index points (full width 15 points, one standard deviation of the reference group). Solid means clearly ahead, hatched not certain yet, a thin grey bar level.

Show 4 more areas that do not count toward the overall index

SearchGPT-6 Sol +1598.6 to 83.3

CybersecurityGPT-6 Sol +15116.4 to 101.8

ProductivityGPT-6 Sol +999.4 to 90.5

Educationonly GPT-6 Luna has a score

Bar length shows the gap in index points (full width 15 points, one standard deviation of the reference group). Solid means clearly ahead, hatched not certain yet, a thin grey bar level.

GPT-6 Sol vs. GPT-6 Luna at the same effort

Both models measured at the same reasoning efforts. This shows whether a lead comes from the model or only from more compute.

maxGPT-6 Sol ahead

GPT-6 Sol 110.5 vs. GPT-6 Luna 96.3 · Gap 16.4 · Test series 11

Overall index (max vs. max)GPT-6 Sol ahead

GPT-6 Sol 110.5 vs. GPT-6 Luna 96.3

Effort	GPT-6 Sol	GPT-6 Luna	Gap	Verdict	Test series
max	110.5	96.3	16.4	GPT-6 Sol ahead	11
Overall index (max vs. max)	110.5	96.3	14.3	GPT-6 Sol ahead

Gap and verdict compare only the Artificial Analysis test series in which both models are measured at that effort. Tied below 3 points, ahead only beyond two standard errors. The single values rest on all test series of each model, so their difference can deviate from the gap.

Which model for which task?

Everyday tasks, translated from the area values above. Each row names the values it rests on.

TaskBetter choicebased on

Writing and debugging codeGPT-6 SolCoding 114.9 to 97.5

Maths and data analysisGPT-6 SolMathematics 117.5 to 105.5, Reasoning 114.1 to 97.8

Facts and researchGPT-6 SolKnowledge 109.8 to 89.2, Search 98.6 to 83.3

Agents and workflowsGPT-6 SolAgents 107.2 to 99.5

Fast chat repliesabout the same111 to 101 tokens per second

Security analysisGPT-6 SolCybersecurity 116.4 to 101.8

Many requests on a small budgetGPT-6 Luna$4 to $0.20 per 1M tokens

Finance, legal and medical textsGPT-6 SolFinance 101.3 to 98.3, Legal 104.8 to 90.7, Health 104.8 to 95.2

Images, screenshots and PDFsGPT-6 SolMultimodal 106.9 to 92.0

See for yourself: same task, one attempt

The same prompt for both models. We show the latest attempt without fixes; it does not count toward the overall index.

Pelican on a bicycle

Prompt

Generate an SVG of a pelican riding a bicycle

Task origin: Simon Willison

GPT-6 Sol

Oct 1, 2026 · 80 s · $0.04

Test details

Run date: Oct 1, 2026
API provider: openrouter
Requested API model: openai/gpt-6-sol
Reasoning setting: high
Token limit including reasoning: 8,192 tokens
Output tokens: 4,389 tokens
Cost of the attempt: $0.0439
Response time: 80.5 s
API model reported in response: openai/gpt-6-sol
Pinned endpoint: OpenAI | openai/gpt-6-sol-20260922
Temperature: Provider default
Test protocol: community-visual-tests-v1
Technical checks: Passed, not a quality rating

Visually reviewed. This individual test does not establish overall model quality.

GPT-6 Luna

Oct 1, 2026 · 47 s · $0.0022

Test details

Run date: Oct 1, 2026
API provider: openrouter
Requested API model: openai/gpt-6-luna
Reasoning setting: high
Token limit including reasoning: 8,192 tokens
Output tokens: 4,493 tokens
Cost of the attempt: $0.0022
Response time: 46.7 s
API model reported in response: openai/gpt-6-luna
Pinned endpoint: OpenAI | openai/gpt-6-luna-20260922
Temperature: Provider default
Test protocol: community-visual-tests-v1
Technical checks: Passed, not a quality rating

Visually reviewed. This individual test does not establish overall model quality.

SVG short test

One SVG task with the same fixed settings for every model, without hidden thinking.

Prompt

Return only one standalone SVG of a pelican riding a bicycle. Use viewBox="0 0 800 600", simple shapes, two clearly visible wheels and a recognizable pelican with a long beak. At most 30 SVG elements and 1800 characters. No Markdown, explanations, comments, scripts, animation, CSS or external resources.

GPT-6 Sol

Unavailable in this round

GPT-6 Luna

Unavailable in this round

GPT-6 Sol and GPT-6 Luna in the overall index

Every model with a place in the overall index, both models of this matchup highlighted. You can add or hide other models.

Overall index (Gradually Meta-Score v4)

20 of 92 models

4 models in this view are measured at a lower reasoning effort than max in the overall index. Dashed column: lower reasoning effort than the top.

Claude Opus 5.5max126.1

Gemini 4 Carbonhigh125.8

Claude Fable 5.1max123.8

Claude Sonnet 5.5max122.8

Gemini 4 Argonhigh122.3

GPT-6 Astramax121.3

GPT-6.1 Solmax119.2

Muse Wonder Fullmax118.3

Claude Opus 5max117.2

Muse Wondermax113.4

Kimi K3.1 Promax112.9

Claude Fable 5max112.5

GPT-6 Solmax110.5

GPT-5.6 Solmax108.7

Muse Spark 1.3max107.4

Gemini 3.7 Flashhigh107.1

DeepSeek V4.1 Flash 1017max106.6

Gemini 3.8 Flashhigh106.5

GPT-6.1 Terramax105.3

GPT-6 Lunamax96.3

Claude Opus 5.5

Gemini 4 Carbon

Claude Fable 5.1

Claude Sonnet 5.5

Gemini 4 Argon

GPT-6 Astra

GPT-6.1 Sol

Muse Wonder Full

Claude Opus 5

Muse Wonder

Kimi K3.1 Pro

Claude Fable 5

GPT-6 Sol

GPT-5.6 Sol

Muse Spark 1.3

Gemini 3.7 Flash

DeepSeek V4.1 Flash 1017

Gemini 3.8 Flash

GPT-6.1 Terra

GPT-6 Luna

Popularity compared

How often the models are searched for, used and mentioned, on a scale from 0 to 100.

GPT-6 Sol is more popular than GPT-6 Luna: 73 to 60 out of 100.

Over time

Weekly values from October 18, 2026 to November 7, 2026

GPT-6 SolGPT-6 Luna

All data side by side

The better value is bold. For the overall index only when the gap is clear.

Metric	GPT-6 Sol	GPT-6 Luna
Intelligence
Overall indexpoints, 100 = reference group average, higher is better	110.5	96.3
Rank	Rank 13 of 92	Rank 35 of 92
Price per 1M tokens
Inputwhat you send	$2above 272,000 tokens: $4	$0.10above 272,000 tokens: $0.20
Outputwhat the model writes	$10above 272,000 tokens: $15	$0.50above 272,000 tokens: $0.75
Cost per task (own tests)average of 3 original-protocol tasks both models passed (SVG short test not included)	$0.044	$0.003
Speed
Tokens per second	111	101
Popularity
Popularitysearch, usage and mentions, 0 to 100	73Rank 3 of 785	60Rank 30 of 785
Technical
Context windowhow much text the model reads at once	1.05 M tokens≈ 1,580 A4 pages	1.05 M tokens≈ 1,580 A4 pages
Knowledge cutoff	Apr 20, 2026	May 18, 2026
Released	Sep 22, 2026	Sep 22, 2026
Availability	Active	Active
Open weights	no	no

Prices without caching or batch discounts · Page count as a rule of thumb from Artificial Analysis

Is there a better choice?

Successor of GPT-6 SolGPT-6.1 Sol

Overall index 119.2, rank 7$2 input and $10 output per 1M tokens

GPT-6 Luna vs. GPT-6.1 Sol

As cheap as GPT-6 LunaClaude Haiku 5.5

Overall index 103.4, rank 20$0.10 input and $0.50 output per 1M tokens

GPT-6 Sol vs. Claude Haiku 5.5

Best model in the overall indexClaude Opus 5.5

Overall index 126.1, rank 1$4 input and $20 output per 1M tokens

GPT-6 Sol vs. Claude Opus 5.5

Shared benchmarks

Single results from the same benchmark version, task and metric. A value is bold only with the same test setup and a clear gap.

A: GPT-6 SolB: GPT-6 Luna

AA-BriefcaseOverall · higher is better1,476.5 elo1,337.1 eloArtificial AnalysisAs of 11/07/2026

AA-OmniscienceOverall · higher is better21.400 points-8.267 pointsArtificial AnalysisAs of 11/07/2026

AA-Omniscience AccuracyOverall · higher is better51.1%39.6%Artificial AnalysisAs of 11/07/2026

AA-Omniscience Hallucination RateOverall · lower is better60.8%79.3%Artificial AnalysisAs of 11/07/2026

AutomationBench-AAOverall · higher is better61.6%53.2%Artificial AnalysisAs of 11/07/2026

BioMysteryBenchOverall · higher is better74.8%61.5%Vals AIAs of 11/05/2026

Code MigrationOverall · higher is better57.2%42.6%Vals AIAs of 10/29/2026

CritPt v2 (Hard Subset)Overall · higher is better66.5%42.5%Artificial AnalysisAs of 11/07/2026

CyberBench v1.1Overall · higher is better78.0%76.3%Vals AIAs of 11/05/2026

EMBOverall · higher is better71.5%68.5%Vals AIAs of 10/20/2026

Show 54 more benchmarks

Finance Agent (v2)Overall · higher is better49.1%49.9%Vals AIAs of 10/29/2026

FrontierCodeOverall · higher is better35.6%25.2%Artificial AnalysisAs of 11/07/2026

GDP.pdfOverall · higher is better25.2%22.8%Artificial AnalysisAs of 11/07/2026

GDPval-AAOverall · higher is better1,515.9 elo1,444.5 eloArtificial AnalysisAs of 11/07/2026

HLE-DiamondOverall · higher is better45.2%22.2%Artificial AnalysisAs of 11/07/2026

IOIOverall · higher is better82.6%55.6%Vals AIAs of 10/29/2026

Legal Research BenchOverall · All-pass · higher is better28.8%30.3%Vals AIAs of 11/05/2026

LMArena Code Arena WebDevOverall · higher is better1,6781,579LMArenaAs of 10/28/2026

LMArena SearchOverall · higher is better1,1751,160LMArenaAs of 10/13/2026

LMArena Text, Style ControlOverall · higher is better1,4581,441LMArenaAs of 10/28/2026

LMArena Vision, Style ControlOverall · higher is better1,2661,240LMArenaAs of 10/19/2026

MedCodeOverall · higher is better47.1%44.7%Vals AIAs of 10/29/2026

MedScribeOverall · higher is better82.0%83.7%Vals AIAs of 10/29/2026

MMMU-ProOverall · higher is better82.9%79.7%Artificial AnalysisAs of 11/07/2026

MysteryMechanismOverall · higher is better30.2%19.4%Vals AIAs of 10/29/2026

ProgramBenchOverall · higher is better2.0%0.5%Vals AIAs of 11/05/2026

ProofBench v1.1Overall · higher is better83%64%Vals AIAs of 10/29/2026

Public Benefits Bench v1.1Overall · higher is better56.6%57.6%Vals AIAs of 10/12/2026

SRE BenchOverall · higher is better30.5%2.7%Vals AIAs of 11/05/2026

Tax Agent BenchOverall · higher is better15.0%19.9%Vals AIAs of 11/05/2026

Terminal-BenchOverall · higher is better43.9%12.6%Artificial AnalysisAs of 11/07/2026

Terminal-Bench 4.0Overall · higher is better44.4%13.6%Vals AIAs of 11/05/2026

Terminal-Bench-ScienceOverall · higher is better30.0%4.3%Vals AIAs of 11/05/2026

Terminal-Bench-ScienceOverall · higher is better30.0%8.6%Artificial AnalysisAs of 11/07/2026

Vals Science IndexOverall · higher is better45.0%28.4%Vals AIAs of 10/25/2026

Vibe Code Bench v1.1Overall · higher is better87.8%81.6%Vals AIAs of 11/05/2026

BioMysteryBenchHard · higher is better35.3%17.6%Vals AIAs of 11/05/2026

Code MigrationCLI · higher is better52.9%37.2%Vals AIAs of 10/29/2026

CyberBench v1.1Patch · higher is better89.3%87.5%Vals AIAs of 11/05/2026

EMBComps · higher is better73.3%70.5%Vals AIAs of 10/20/2026

Finance Agent (v2)Adjustments · higher is better43.2%49.8%Vals AIAs of 10/29/2026

IOIIOI 2024 · higher is better100.0%69.3%Vals AIAs of 10/29/2026

Legal Research BenchAdministrative / Regulatory · All-pass · higher is better39.4%37.9%Vals AIAs of 11/05/2026

LMArena Code Arena WebDevHTML development · higher is better1,6801,595LMArenaAs of 10/28/2026

LMArena Text, Style ControlBusiness, management and finance · higher is better1,4481,441LMArenaAs of 10/28/2026

LMArena Vision, Style ControlCreative writing with images · higher is better1,2521,232LMArenaAs of 10/19/2026

ProgramBenchAlmost Resolved · higher is better32.5%8.0%Vals AIAs of 11/05/2026

SRE BenchCapability Score · higher is better57.3%11.1%Vals AIAs of 11/05/2026

Tax Agent BenchControversy & Precedence Analysis · higher is better6.7%6.7%Vals AIAs of 11/05/2026

Terminal-Bench 4.0Hardware · higher is better6.7%0.0%Vals AIAs of 11/05/2026

Terminal-Bench-ScienceEarth Sciences · higher is better25%0%Vals AIAs of 11/05/2026

Vals Science IndexBioMysteryBench · higher is better74.8%61.5%Vals AIAs of 10/25/2026

BioMysteryBenchHuman-solvable · higher is better84.0%71.7%Vals AIAs of 11/05/2026

Code MigrationCOBOL · higher is better70.0%58.6%Vals AIAs of 10/29/2026

CyberBench v1.1PoC · higher is better66.7%65.0%Vals AIAs of 11/05/2026

EMBDataroom · higher is better83.9%75.5%Vals AIAs of 10/20/2026

Finance Agent (v2)All-Pass · higher is better36.0%38.4%Vals AIAs of 10/29/2026

IOIIOI 2025 · higher is better79.3%39.0%Vals AIAs of 10/29/2026

Legal Research BenchAdministrative / Regulatory · Weighted partial credit (weighted pass) · higher is better80.5%83.2%Vals AIAs of 11/05/2026

LMArena Code Arena WebDevImage to WebDev · higher is better1,6601,571LMArenaAs of 10/19/2026

LMArena Text, Style ControlChinese · higher is better1,4851,483LMArenaAs of 10/28/2026

LMArena Vision, Style ControlDiagram understanding · higher is better1,3031,275LMArenaAs of 10/19/2026

ProgramBenchFully Resolved · higher is better2.0%0.5%Vals AIAs of 11/05/2026

SRE BenchFully Solved · higher is better30.5%2.7%Vals AIAs of 11/05/2026

More comparisons

Compare GPT-6 Sol and GPT-6 Luna with the models people compare them with most often.

GPT-6 Solvs.GPT-6 Luna

Choose GPT-6 Sol for

Choose GPT-6 Luna for

Where is the difference?

Strengths at a glance

GPT-6 Sol vs. GPT-6 Luna at the same effort

Which model for which task?

See for yourself: same task, one attempt

Pelican on a bicycle

SVG short test

GPT-6 Sol

GPT-6 Luna

GPT-6 Sol and GPT-6 Luna in the overall index

Overall index (Gradually Meta-Score v4)

Popularity compared

Over time

All data side by side

Is there a better choice?

Shared benchmarks

More comparisons

Compare another matchup

Predecessor and successor

The most popular comparisons