Enter opens the highlighted result. Esc closes.

GPT-5.5 vs GPT-5.6 Luna

GPT-5.5 leads GPT-5.6 Luna on coding, 71.3 to 58.0 on the Vibe Coding Index, and GPT-5.6 Luna costs 25.0x less per output token ($1.20 vs $30.00 per million). Updated October 22, 2026; 109 benchmarks have a result for both.

Replace GPT-5.5 Replace GPT-5.6 Luna
GPT-5.5OpenAI
Coding
71.3
Overall
69.2
Output $/M
$30.00
92
GPT-5.6 LunaOpenAI
Coding
58.0
Overall
61.6
Output $/M
$1.20
17
vs boards won of 109 shared

GPT-5.5 or GPT-5.6 Luna: which to pick

Pick GPT-5.5 for coding: it scores 71.3 on the Vibe Coding Index against 58.0 for GPT-5.6 Luna, though on 5 benchmarks against 6.

Pick GPT-5.6 Luna for output tokens 25.0x cheaper ($1.20 vs $30.00 per million) and a lower cost per solved task ($0.04 against $1.65).

Model AGPT-5.5OpenAI71.3/1005 independent benchmarks$1.65 per solved taskModel BGPT-5.6 LunaOpenAI58.0/1006 independent benchmarks$0.04 per solved task
Shared results
109
GPT-5.5 only
25
GPT-5.6 Luna only
10
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-5.5 71.3 from 5 benchmarks, GPT-5.6 Luna 58.0 from 6 benchmarks.
Widest category
Reasoning, 11.8 points apart on average over 17 shared benchmarks.
Widest single gap
Kagi LLM Benchmark: GPT-5.5 88.8% against GPT-5.6 Luna 49.1%, 39.7 points apart on a 0 to 100 scale.
Output price
GPT-5.5 $30.00 against GPT-5.6 Luna $1.20 per million output tokens, 25.0x 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-5.5BarsGPT-5.6 LunaTies
Coding135
Intelligence20
Agentic work65
Reasoning161
Math42
Knowledge111
Web design81
General322
All shared boards9217
CodingLike-for-likeSix of these boards feed the Vibe Coding Index135
Coding benchmarks for GPT-5.5 and GPT-5.6 Luna
BenchmarkGPT-5.5GPT-5.6 LunaLeader
LiveBench coding85.1%73.6%GPT-5.5
LMArena WebDev15131518GPT-5.6 Luna
AA Coding Index74.971.4GPT-5.5
LiveCodeBench (Vals AI)85.3%Not measuredOnly GPT-5.5 measured
SWE-bench Verified (Vals AI)82.6%93.0%GPT-5.6 Luna
Terminal-Bench 4 (Vals AI)Not measured11.6%Only GPT-5.6 Luna measured
Vibe Code Bench (Vals AI)69.8%77.1%GPT-5.6 Luna
Code migration (Vals AI)45.2%44.5%GPT-5.5
ProgramBench (Vals AI)0.5%0.0%GPT-5.5
IOI (Vals AI)Not measured61.8%Only GPT-5.6 Luna measured
Terminal-Bench 2.1 (Vals AI)76.4%79.0%GPT-5.6 Luna
Vibe Code Bench 1-100 (Vals AI)Not measured22.6%Only GPT-5.6 Luna measured
SkillsBench (Vals AI)62.5%60.5%GPT-5.5
CyberBench (Vals AI)Not measured73.6%Only GPT-5.6 Luna measured
SciCode56.1%53.6%GPT-5.5
Terminal-Bench Hard (Artificial Analysis run)60.6%Not measuredOnly GPT-5.5 measured
Terminal-Bench84.7%Not measuredOnly GPT-5.5 measured
FrontierCode43.0%39.8%GPT-5.5
GSO-Bench40.2%Not measuredOnly GPT-5.5 measured
WeirdML84.9%60.9%GPT-5.5
MirrorCode10.0%Not measuredOnly GPT-5.5 measured
ALE-Bench1943.01667.4GPT-5.5
LMArena Text: coding14941466GPT-5.5
LMArena Text: software and IT14931463GPT-5.5
LMArena WebDev: React15061508GPT-5.6 Luna
LMArena WebDev: HTML15431520GPT-5.5
LMArena WebDev: from an image15321517GPT-5.5
SciCodeNot measured6.2%Only GPT-5.6 Luna measured
Neither model has a result on the other 11 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for GPT-5.5 and GPT-5.6 Luna
BenchmarkGPT-5.5GPT-5.6 LunaLeader
AA Intelligence Index38.437.3GPT-5.5
Epoch Capabilities Index159.7156.5GPT-5.5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index65
Agentic work benchmarks for GPT-5.5 and GPT-5.6 Luna
BenchmarkGPT-5.5GPT-5.6 LunaLeader
AA Agentic Index36.442.1GPT-5.6 Luna
Design Arena: agents building web apps1117Not measuredOnly GPT-5.5 measured
Design Arena: agents building full-stack apps1089Not measuredOnly GPT-5.5 measured
Design Arena: agents building mobile apps1142Not measuredOnly GPT-5.5 measured
Design Arena: agents building Android apps1164Not measuredOnly GPT-5.5 measured
Design Arena: agents building Godot games1173Not measuredOnly GPT-5.5 measured
Design Arena: agents building games1179Not measuredOnly GPT-5.5 measured
Design Arena: HTML slides1017Not measuredOnly GPT-5.5 measured
Design Arena: PowerPoint slides1084Not measuredOnly GPT-5.5 measured
LiveBench research61.6%51.0%GPT-5.5
LiveBench browser48.0%68.0%GPT-5.6 Luna
LiveBench long agentic44.6%30.4%GPT-5.5
SREBench (Vals AI)3.8%1.5%GPT-5.5
Finance Agent (Vals AI)51.8%55.0%GPT-5.6 Luna
Tax Agent Bench (Vals AI)21.8%24.8%GPT-5.6 Luna
Legal Research Bench (Vals AI)40.4%36.5%GPT-5.5
Excel Modeling Benchmark (Vals AI)64.5%67.1%GPT-5.6 Luna
BioMysteryBench (Vals AI)Not measured61.5%Only GPT-5.6 Luna measured
Terminal-Bench-Science (Vals AI)Not measured0.0%Only GPT-5.6 Luna measured
Vals Science IndexNot measured25.3%Only GPT-5.6 Luna measured
Time Horizon Index (Vals AI)8.3%Not measuredOnly GPT-5.5 measured
τ²-Bench Telecom93.9%Not measuredOnly GPT-5.5 measured
APEX-Agents55.1%43.0%GPT-5.5
Remote Labor Index6.3%Not measuredOnly GPT-5.5 measured
OSWorld 2.013.0%Not measuredOnly GPT-5.5 measured
BALROGNot measured45.6%Only GPT-5.6 Luna measured
DeepResearch Bench54.0%Not measuredOnly GPT-5.5 measured
PostTrainBench27.2%Not measuredOnly GPT-5.5 measured
Vending-Bench 2$7,524$4,095GPT-5.5
Neither model has a result on the other 12 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index161
Reasoning benchmarks for GPT-5.5 and GPT-5.6 Luna
BenchmarkGPT-5.5GPT-5.6 LunaLeader
GPQA Diamond93.2%91.7%GPT-5.5
LiveBench reasoning89.7%85.6%GPT-5.5
MysteryMechanism (Vals AI)Not measured14.4%Only GPT-5.6 Luna measured
MMMU Pro (Vals AI)88.3%85.0%GPT-5.5
GPQA Diamond (Artificial Analysis run)93.5%91.1%GPT-5.5
Humanity's Last Exam (Artificial Analysis run)45.8%39.5%GPT-5.5
GPQA Diamond (Epoch AI run)90.7%91.6%GPT-5.6 Luna
ARC-AGI95.0%88.0%GPT-5.5
ARC-AGI-285.0%59.5%GPT-5.5
SimpleBench69.0%46.8%GPT-5.5
Chess puzzles63.0%43.0%GPT-5.5
Mystery game puzzles56.0%21.0%GPT-5.5
EBR-bench34.3%Not measuredOnly GPT-5.5 measured
LMArena Text: hard prompts14891451GPT-5.5
LMArena Text: expert15071478GPT-5.5
LMArena Text: hard prompts in English14831460GPT-5.5
LMArena Vision: diagrams13251273GPT-5.5
LMArena Vision: homework13321282GPT-5.5
Kagi LLM Benchmark88.8%49.1%GPT-5.5
Neither model has a result on the other 3 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index42
Math benchmarks for GPT-5.5 and GPT-5.6 Luna
BenchmarkGPT-5.5GPT-5.6 LunaLeader
OTIS Mock AIME84.4%98.3%GPT-5.6 Luna
FrontierMath, tiers 1 to 385.3%82.1%GPT-5.5
FrontierMath, tier 472.5%61.0%GPT-5.5
ProofBench50.0%60.0%GPT-5.6 Luna
LMArena Text: math14871459GPT-5.5
LMArena Text: mathematical work14961470GPT-5.5
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index111
Knowledge benchmarks for GPT-5.5 and GPT-5.6 Luna
BenchmarkGPT-5.5GPT-5.6 LunaLeader
MMLU Pro88.1%86.0%GPT-5.5
LegalBench (Vals AI)86.5%84.0%GPT-5.5
CaseLaw (Vals AI)66.2%Not measuredOnly GPT-5.5 measured
CorpFin (Vals AI)68.4%64.2%GPT-5.5
TaxEval (Vals AI)75.0%76.2%GPT-5.6 Luna
MortgageTax (Vals AI)68.8%67.3%GPT-5.5
MedCode (Vals AI)49.1%42.4%GPT-5.5
SimpleQA Verified63.0%41.0%GPT-5.5
LMArena Text: science14851433GPT-5.5
LMArena Text: medicine and healthcare14651399GPT-5.5
LMArena Text: legal and government14911446GPT-5.5
LMArena Text: business and finance14781439GPT-5.5
FACTS Benchmark Suite66.7%35.0%GPT-5.5
Neither model has a result on the other 5 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index81
Web design benchmarks for GPT-5.5 and GPT-5.6 Luna
BenchmarkGPT-5.5GPT-5.6 LunaLeader
Design Arena: websites12561255GPT-5.5
Design Arena: UI components12571248GPT-5.5
Design Arena: game dev12931194GPT-5.5
Design Arena: data viz12641211GPT-5.5
Design Arena: 3D11971121GPT-5.5
Design Arena: all code categories12601231GPT-5.5
Design Arena: SVG12341194GPT-5.5
Design Arena: ASCII art12551142GPT-5.5
LiveBench design70.3%76.0%GPT-5.6 Luna
GeneralLike-for-likePublic boards, shown as published, kept out of the index322
General benchmarks for GPT-5.5 and GPT-5.6 Luna
BenchmarkGPT-5.5GPT-5.6 LunaLeader
LiveBench data analysis81.6%78.0%GPT-5.5
Public Benefits Bench (Vals AI)60.9%61.2%GPT-5.6 Luna
SAGE (Vals AI)51.5%44.2%GPT-5.5
MedScribe (Vals AI)86.9%84.4%GPT-5.5
Vals Multimodal Index68.1%69.1%GPT-5.6 Luna
IFBench75.9%Not measuredOnly GPT-5.5 measured
Long context reasoning (AA-LCR)84.3%83.7%GPT-5.5
DTBench96.0%89.1%GPT-5.5
LMCA54.3%48.5%GPT-5.5
Surface Evolver Bench88.1%61.9%GPT-5.5
Furniture assembly44.2%42.5%GPT-5.5
ExploitBench47.4%Not measuredOnly GPT-5.5 measured
CL-bench Life22.2%Not measuredOnly GPT-5.5 measured
LMArena Text14711431GPT-5.5
LMArena Text: instruction following14781437GPT-5.5
LMArena Text: creative writing14551396GPT-5.5
LMArena Vision12971257GPT-5.5
LMArena Text: English14701441GPT-5.5
LMArena Text: non-English14671417GPT-5.5
LMArena Text: Chinese15341470GPT-5.5
LMArena Text: German14801455GPT-5.5
LMArena Text: Japanese14991415GPT-5.5
LMArena Text: Korean14601415GPT-5.5
LMArena Text: French14851458GPT-5.5
LMArena Text: Spanish14681446GPT-5.5
LMArena Text: Russian14731427GPT-5.5
LMArena Text: Polish14651405GPT-5.5
LMArena Text: multi-turn14761434GPT-5.5
LMArena Text: longer queries14841436GPT-5.5
LMArena Text: writing and language14661419GPT-5.5
LMArena Text: entertainment and media14491402GPT-5.5
LMArena Vision: English12921263GPT-5.5
LMArena Vision: Chinese13601294GPT-5.5
LMArena Vision: reading text13061264GPT-5.5
LMArena Vision: humor12931190GPT-5.5
LMArena Vision: recognizing things1231Not measuredOnly GPT-5.5 measured
LMArena Vision: creative writing13011257GPT-5.5
LMArena Document14861457GPT-5.5
Neither model has a result on the other 8 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-5.5 and GPT-5.6 Luna
BenchmarkGPT-5.5GPT-5.6 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

GPT-5.5$0.020Fits in one request
GPT-5.6 Luna$0.0008Fits in one request

GPT-5.6 Luna costs less

Repository review

50K input + 3K output tokens

GPT-5.5$0.340Fits in one request
GPT-5.6 Luna$0.014Fits in one request

GPT-5.6 Luna costs less

Agent loop

200K input + 10K output tokens

GPT-5.5$1.30Fits in one request
GPT-5.6 Luna$0.052Fits in one request

GPT-5.6 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-5.5 and GPT-5.6 Luna
SpecificationOpenAI GPT-5.5OpenAI GPT-5.6 Luna
MakerOpenAIOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedApril 24, 2026July 9, 2026
Context window1.05M1.05M
Input $/Mtok$5.00$0.20
Output $/Mtok$30.00$1.20
Coding index71.358.0
Independent measurements behind it56
Cost per solved task$1.65$0.04
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-5.5 better than GPT-5.6 Luna for coding?

Pick GPT-5.5 for coding: it scores 71.3 on the Vibe Coding Index against 58.0 for GPT-5.6 Luna, though on 5 benchmarks against 6. Pick GPT-5.6 Luna for output tokens 25.0x cheaper ($1.20 vs $30.00 per million) and a lower cost per solved task ($0.04 against $1.65).

Which is cheaper, GPT-5.5 or GPT-5.6 Luna?

GPT-5.5 costs $5.00 per million input tokens and $30.00 per million output tokens; GPT-5.6 Luna costs $0.20 and $1.20. Per solved SWE-bench Verified task, GPT-5.5 costs $1.65 and GPT-5.6 Luna $0.04.

Which has the bigger context window, GPT-5.5 or GPT-5.6 Luna?

Neither: both take 1.05M tokens.

How many benchmarks is the GPT-5.5 vs GPT-5.6 Luna comparison based on?

109 benchmarks have a result for both models as of October 22, 2026. GPT-5.5's coding index rests on 5 independent benchmarks and GPT-5.6 Luna's on 6. 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-5-5-vs-gpt-5-6-luna" title="GPT-5.5 vs GPT-5.6 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/gpt-5-5-vs-gpt-5-6-luna">LLMBenchmark.ai, GPT-5.5 vs GPT-5.6 Luna</a></p>