GPT-5.4 vs GPT-6 Luna

Updated October 21, 2026. 75 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace GPT-5.4 Replace GPT-6 Luna

GPT-5.4OpenAI

Coding: 37.7
Overall: 39.0
Output $/M: $15.00

GPT-6 LunaOpenAI

Coding: 71.4
Overall: 38.1
Output $/M: $0.50

vs boards won of 75 shared, 1 tied

Decision reading

Pick GPT-6 Luna for coding: it scores 71.4 on the Vibe Coding Index against 37.7 for GPT-5.4.

Pick GPT-5.4 when what matters is ALE-Bench or LMArena Text: coding, where it leads.

Model AGPT-5.4OpenAI37.7/1001 independent benchmark$1.02 per solved task Model BGPT-6 LunaOpenAI71.4/1002 independent benchmarksNo cost per solved task

753427

Shared results: 75
GPT-5.4 only: 34
GPT-6 Luna only: 27
Like-for-like categories: 8 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

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.

Coding
Vibe Coding Index, four public coding benchmarks
GPT-6 Luna
GPT-6 Luna scores 71.4 against 37.7.
1 and 2 benchmarks behind the scores
Intelligence
AA Intelligence Index
GPT-5.4
GPT-5.4 leads, 39.0 against 38.1.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only GPT-5.4 has a GPQA Diamond result, 91.7%.
Directional only
Math
AIME
No call
Only GPT-5.4 has a AIME result, 96.7%.
Directional only
Knowledge
MMLU Pro
No call
Only GPT-5.4 has a MMLU Pro result, 87.5%.
Directional only
Web design
Design Arena: websites
GPT-6 Luna
GPT-6 Luna leads, 1272 against 1199.
Same board, both measured
General
Vals Index
No call
Only GPT-6 Luna has a Vals Index result, 52.3%.
Directional only
Long documents
Listed context window
No call
Both take 1.05M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GPT-6 Luna
GPT-5.4 $0.010, GPT-6 Luna $0.0003 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only GPT-5.4 has the number, $1.02.
Directional only

At a glance

Coding index: GPT-5.4 37.7 from 1 benchmark, GPT-6 Luna 71.4 from 2 benchmarks.
Widest category: Coding, 12.2 points apart on average over 10 shared benchmarks.
Widest single gap: Mystery game puzzles: GPT-5.4 37.0% against GPT-6 Luna 7.0%, 30.0 points apart on a 0 to 100 scale.
Output price: GPT-5.4 $15.00 against GPT-6 Luna $0.50 per million output tokens, 30.0x 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
Category	GPT-5.4	GPT-6 Luna	Ties
Coding	3	6	1 tie
Intelligence	2	0
Agentic work	1	3
Reasoning	11	0
Math	2	4
Knowledge	5	1
Web design	1	7
General	24	4
All shared boards	49	25	1 tie

CodingLike-for-likeFour of these boards feed the Vibe Coding Index36

Coding benchmarks for GPT-5.4 and GPT-6 Luna
Benchmark	GPT-5.4	GPT-6 Luna	Leader
LiveBench coding	Not measured	84.5%	Only GPT-6 Luna measured
LMArena WebDev	1464	1578	GPT-6 Luna
AA Coding Index	71.1	Not measured	Only GPT-5.4 measured
LiveCodeBench (Vals AI)	84.1%	Not measured	Only GPT-5.4 measured
SWE-bench Verified (Vals AI)	78.2%	Not measured	Only GPT-5.4 measured
Terminal-Bench 4 (Vals AI)	Not measured	13.6%	Only GPT-6 Luna measured
Vibe Code Bench (Vals AI)	67.4%	81.7%	GPT-6 Luna
Code migration (Vals AI)	35.0%	42.5%	GPT-6 Luna
ProgramBench (Vals AI)	0.5%	0.5%	Tie
IOI (Vals AI)	Not measured	55.6%	Only GPT-6 Luna measured
Terminal-Bench 2.1 (Vals AI)	Not measured	73.0%	Only GPT-6 Luna measured
SWE-bench Multimodal (Vals AI)	Not measured	37.5%	Only GPT-6 Luna measured
SkillsBench (Vals AI)	51.7%	Not measured	Only GPT-5.4 measured
CyberBench (Vals AI)	Not measured	76.3%	Only GPT-6 Luna measured
SciCode	Not measured	54.6%	Only GPT-6 Luna measured
Terminal-Bench Hard (Artificial Analysis run)	57.6%	Not measured	Only GPT-5.4 measured
Terminal-Bench	81.8%	Not measured	Only GPT-5.4 measured
FrontierCode	Not measured	42.4%	Only GPT-6 Luna measured
GSO-Bench	31.4%	Not measured	Only GPT-5.4 measured
WeirdML	77.7%	Not measured	Only GPT-5.4 measured
MirrorCode	15.6%	Not measured	Only GPT-5.4 measured
ALE-Bench	1607.0	1576.9	GPT-5.4
LMArena Text: coding	1496	1437	GPT-5.4
LMArena Text: software and IT	1491	1423	GPT-5.4
LMArena WebDev: React	1454	1573	GPT-6 Luna
LMArena WebDev: HTML	1478	1595	GPT-6 Luna
LMArena WebDev: from an image	1436	1571	GPT-6 Luna
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-5.4 and GPT-6 Luna
Benchmark	GPT-5.4	GPT-6 Luna	Leader
AA Intelligence Index	39.0	38.1	GPT-5.4
Epoch Capabilities Index	157.3	156.8	GPT-5.4

Agentic workLike-for-likePublic boards, shown as published, kept out of the index13

Agentic work benchmarks for GPT-5.4 and GPT-6 Luna
Benchmark	GPT-5.4	GPT-6 Luna	Leader
Design Arena: agents building web apps	1059	1189	GPT-6 Luna
Design Arena: agents building full-stack apps	1017	1158	GPT-6 Luna
Design Arena: agents building mobile apps	1087	1133	GPT-6 Luna
Design Arena: agents building Android apps	1037	Not measured	Only GPT-5.4 measured
Design Arena: agents building Godot games	1135	Not measured	Only GPT-5.4 measured
LiveBench research	Not measured	43.5%	Only GPT-6 Luna measured
LiveBench browser	Not measured	72.0%	Only GPT-6 Luna measured
LiveBench long agentic	Not measured	48.1%	Only GPT-6 Luna measured
SREBench (Vals AI)	Not measured	2.7%	Only GPT-6 Luna measured
Finance Agent (Vals AI)	Not measured	49.9%	Only GPT-6 Luna measured
Tax Agent Bench (Vals AI)	Not measured	19.9%	Only GPT-6 Luna measured
Legal Research Bench (Vals AI)	Not measured	30.3%	Only GPT-6 Luna measured
Harvey's Legal Agent Benchmark (Vals AI)	Not measured	4.2%	Only GPT-6 Luna measured
Excel Modeling Benchmark (Vals AI)	Not measured	68.5%	Only GPT-6 Luna measured
BioMysteryBench (Vals AI)	Not measured	61.5%	Only GPT-6 Luna measured
Terminal-Bench-Science (Vals AI)	Not measured	4.3%	Only GPT-6 Luna measured
Vals Science Index	Not measured	28.4%	Only GPT-6 Luna measured
τ²-Bench Telecom	87.1%	Not measured	Only GPT-5.4 measured
APEX-Agents	52.4%	44.3%	GPT-5.4
METR time horizons	74.3%	Not measured	Only GPT-5.4 measured
DeepResearch Bench	35.1%	Not measured	Only GPT-5.4 measured
PostTrainBench	19.0%	Not measured	Only GPT-5.4 measured
SpireClimb	Not measured	0.0%	Only GPT-6 Luna measured
Vending-Bench 2	$6,144	Not measured	Only GPT-5.4 measured
Neither model has a result on the other 15 boards here

ReasoningLike-for-likePublic boards, shown as published, kept out of the index110

Reasoning benchmarks for GPT-5.4 and GPT-6 Luna
Benchmark	GPT-5.4	GPT-6 Luna	Leader
GPQA Diamond	91.7%	Not measured	Only GPT-5.4 measured
LiveBench reasoning	Not measured	81.8%	Only GPT-6 Luna measured
MysteryMechanism (Vals AI)	Not measured	19.4%	Only GPT-6 Luna measured
MMMU Pro (Vals AI)	87.5%	Not measured	Only GPT-5.4 measured
GPQA Diamond (Artificial Analysis run)	92.0%	Not measured	Only GPT-5.4 measured
Humanity's Last Exam (Artificial Analysis run)	43.7%	38.5%	GPT-5.4
GPQA Diamond (Epoch AI run)	93.3%	90.5%	GPT-5.4
ARC-AGI	93.7%	86.7%	GPT-5.4
ARC-AGI-2	74.0%	59.3%	GPT-5.4
Chess puzzles	48.0%	36.0%	GPT-5.4
Mystery game puzzles	37.0%	7.0%	GPT-5.4
EBR-bench	25.4%	Not measured	Only GPT-5.4 measured
LMArena Text: hard prompts	1485	1410	GPT-5.4
LMArena Text: expert	1507	1442	GPT-5.4
LMArena Text: hard prompts in English	1479	1410	GPT-5.4
LMArena Vision: diagrams	1327	1245	GPT-5.4
LMArena Vision: homework	1336	1261	GPT-5.4
Neither model has a result on the other 4 boards here

MathLike-for-likePublic boards, shown as published, kept out of the index24

Math benchmarks for GPT-5.4 and GPT-6 Luna
Benchmark	GPT-5.4	GPT-6 Luna	Leader
AIME	96.7%	Not measured	Only GPT-5.4 measured
OTIS Mock AIME	97.8%	98.9%	GPT-6 Luna
FrontierMath, tiers 1 to 3	78.6%	78.9%	GPT-6 Luna
FrontierMath, tier 4	49.0%	56.1%	GPT-6 Luna
ProofBench	56.0%	64.0%	GPT-6 Luna
LMArena Text: math	1488	1409	GPT-5.4
LMArena Text: mathematical work	1497	1427	GPT-5.4
Neither model has a result on the other 6 boards here

KnowledgeLike-for-likePublic boards, shown as published, kept out of the index51

Knowledge benchmarks for GPT-5.4 and GPT-6 Luna
Benchmark	GPT-5.4	GPT-6 Luna	Leader
MMLU Pro	87.5%	Not measured	Only GPT-5.4 measured
LegalBench (Vals AI)	86.0%	Not measured	Only GPT-5.4 measured
CaseLaw (Vals AI)	63.8%	Not measured	Only GPT-5.4 measured
CorpFin (Vals AI)	65.3%	Not measured	Only GPT-5.4 measured
TaxEval (Vals AI)	74.0%	Not measured	Only GPT-5.4 measured
MortgageTax (Vals AI)	68.3%	Not measured	Only GPT-5.4 measured
MedQA (Vals AI)	96.1%	Not measured	Only GPT-5.4 measured
MedCode (Vals AI)	41.3%	44.7%	GPT-6 Luna
SimpleQA Verified	45.1%	41.4%	GPT-5.4
LMArena Text: science	1479	1386	GPT-5.4
LMArena Text: medicine and healthcare	1458	1352	GPT-5.4
LMArena Text: legal and government	1486	1400	GPT-5.4
LMArena Text: business and finance	1474	1388	GPT-5.4
Neither model has a result on the other 4 boards here

Web designLike-for-likePublic boards, shown as published, kept out of the index17

Web design benchmarks for GPT-5.4 and GPT-6 Luna
Benchmark	GPT-5.4	GPT-6 Luna	Leader
Design Arena: websites	1199	1272	GPT-6 Luna
Design Arena: UI components	1215	1297	GPT-6 Luna
Design Arena: game dev	1195	1259	GPT-6 Luna
Design Arena: data viz	1196	1215	GPT-6 Luna
Design Arena: 3D	1065	1217	GPT-6 Luna
Design Arena: all code categories	1180	1271	GPT-6 Luna
Design Arena: SVG	1141	1253	GPT-6 Luna
Design Arena: ASCII art	1188	1156	GPT-5.4
LiveBench design	Not measured	82.5%	Only GPT-6 Luna measured

GeneralLike-for-likePublic boards, shown as published, kept out of the index244

General benchmarks for GPT-5.4 and GPT-6 Luna
Benchmark	GPT-5.4	GPT-6 Luna	Leader
Vals Index	Not measured	52.3%	Only GPT-6 Luna measured
LiveBench data analysis	Not measured	73.4%	Only GPT-6 Luna measured
Public Benefits Bench (Vals AI)	Not measured	57.6%	Only GPT-6 Luna measured
SAGE (Vals AI)	43.3%	48.1%	GPT-6 Luna
MedScribe (Vals AI)	77.5%	83.7%	GPT-6 Luna
IFBench	73.9%	Not measured	Only GPT-5.4 measured
Long context reasoning (AA-LCR)	82.0%	83.3%	GPT-6 Luna
DTBench	94.4%	90.1%	GPT-5.4
LMCA	52.0%	44.5%	GPT-5.4
CL-bench	27.9%	Not measured	Only GPT-5.4 measured
Furniture assembly	37.5%	44.2%	GPT-6 Luna
CL-bench Life	21.7%	Not measured	Only GPT-5.4 measured
LMArena Text	1469	1389	GPT-5.4
LMArena Text: instruction following	1469	1408	GPT-5.4
LMArena Text: creative writing	1439	1363	GPT-5.4
LMArena Vision	1303	1216	GPT-5.4
LMArena Text: English	1467	1393	GPT-5.4
LMArena Text: non-English	1465	1381	GPT-5.4
LMArena Text: Chinese	1520	1427	GPT-5.4
LMArena Text: German	1472	1376	GPT-5.4
LMArena Text: Japanese	1485	1369	GPT-5.4
LMArena Text: Korean	1449	1359	GPT-5.4
LMArena Text: French	1493	1408	GPT-5.4
LMArena Text: Spanish	1452	1391	GPT-5.4
LMArena Text: Russian	1480	1392	GPT-5.4
LMArena Text: Polish	1472	1383	GPT-5.4
LMArena Text: multi-turn	1482	1392	GPT-5.4
LMArena Text: longer queries	1473	1408	GPT-5.4
LMArena Text: writing and language	1458	1384	GPT-5.4
LMArena Text: entertainment and media	1438	1364	GPT-5.4
LMArena Vision: English	1301	1204	GPT-5.4
LMArena Vision: Chinese	1354	1266	GPT-5.4
LMArena Vision: reading text	1312	1225	GPT-5.4
LMArena Vision: humor	1284	Not measured	Only GPT-5.4 measured
LMArena Vision: recognizing things	1251	Not measured	Only GPT-5.4 measured
LMArena Vision: creative writing	1283	1205	GPT-5.4
LMArena Document	1471	Not measured	Only GPT-5.4 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 GPT-5.4 and GPT-6 Luna
Benchmark	GPT-5.4	GPT-6 Luna	Leader
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-5.4$0.010Fits in one request

GPT-6 Luna$0.0003Fits in one request

GPT-6 Luna costs less

Repository review

50K input + 3K output tokens

GPT-5.4$0.170Fits in one request

GPT-6 Luna$0.0065Fits in one request

GPT-6 Luna costs less

Agent loop

200K input + 10K output tokens

GPT-5.4$0.650Fits 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 GPT-5.4 and GPT-6 Luna
Specification	OpenAI GPT-5.4	OpenAI GPT-6 Luna
Maker	OpenAI	OpenAI
Weights	No public weights listed	No public weights listed
Released	March 5, 2026	September 22, 2026
Context window	1.05M	1.05M
Input $/Mtok	$2.50	$0.10
Output $/Mtok	$15.00	$0.50
Coding index	37.7	71.4
Independent measurements behind it	1	2
Cost per solved task	$1.02	Not computable
Source	OpenRouter listing	OpenRouter listing

Questions

Is GPT-5.4 better than GPT-6 Luna for coding?

Pick GPT-6 Luna for coding: it scores 71.4 on the Vibe Coding Index against 37.7 for GPT-5.4. Pick GPT-5.4 when what matters is ALE-Bench or LMArena Text: coding, where it leads.

Which is cheaper, GPT-5.4 or GPT-6 Luna?

GPT-5.4 costs $2.50 per million input tokens and $15.00 per million output tokens; GPT-6 Luna costs $0.10 and $0.50. Cost per solved task is known for GPT-5.4 only.

Which has the bigger context window, GPT-5.4 or GPT-6 Luna?

Neither: both take 1.05M tokens.

How many benchmarks is the GPT-5.4 vs GPT-6 Luna comparison based on?

75 benchmarks have a result for both models as of October 21, 2026. GPT-5.4's coding index rests on 1 independent benchmark and GPT-6 Luna's on 2. Where only one model has a result, the page says so instead of naming a winner.

Decision reading

Which one for your work

Coding

Intelligence

Reasoning

Math

Knowledge

Web design

General

Long documents

Everyday cost

Cost per solved task

At a glance

Benchmark by benchmark

What each workload costs

Chat turn

Repository review

Agent loop

Specification differences

Questions

Compare next