Noometry

Model comparison

GPT-4.5 vs GPT-6 Luna

GPT-6 Luna is the stronger model overall, scoring 53.3 to 37.2 on the Noometry Index.

Last verified . 22 shared benchmarks.

GPT-4.5 OpenAI

37.2

Rank #208 Confirmed

GPT-6 Luna OpenAI

53.3

Rank #36 Confirmed

Summary

  • They share 22 benchmarks with published results for both. GPT-4.5 scores higher in 1 category and GPT-6 Luna in 9 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-6 Luna leads 76.1 to 32.6.
  • The biggest single-benchmark swing is ARC-AGI-1: 10.3% for GPT-4.5 and 86.7% for GPT-6 Luna.

Side by side

GPT-4.5GPT-6 Luna
ProviderOpenAIOpenAI
Noometry Index37.253.3
Released2025-02-272026-09-22
WeightsProprietaryProprietary
Context window—1.05M
Max output—128K
Input $ / M tokens—$0.10
Output $ / M tokens—$0.50
Results tracked4242

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding GPT-6 Luna leads

GPT-4.5: 42.2 (#109), GPT-6 Luna: 55.5 (#25)

BenchmarkGPT-4.5GPT-6 Luna
LMArena Coding13961439
DeepSWE—66.6%
FrontierCode—42.4%
Aider Polyglot44.9%—
LMArena WebDev—1581
SciCode—54.6%
WeirdML39.4%—
LiveBench Coding75.2%—
ALE-Bench—1,577

Agentic & Tool Use GPT-6 Luna leads

GPT-4.5: 27.9 (#97), GPT-6 Luna: 33.3 (#54)

BenchmarkGPT-4.5GPT-6 Luna
APEX-Agents—44.3%
Cybench17.5%—
GDP.pdf—23%

Reasoning GPT-6 Luna leads

GPT-4.5: 13.9 (#330), GPT-6 Luna: 48.2 (#41)

BenchmarkGPT-4.5GPT-6 Luna
ARC-AGI-20.8%59.3%
ARC-AGI-110.3%86.7%
LMArena Hard Prompts14031411
Epoch Capabilities Index136.74156.28
SimpleBench34.5%—
NYT Connections (extended)—68.7%
CritPt—19.4%
Chess Puzzles—31%
EnigmaEval3.2%—
LiveBench Reasoning71.1%—
Mystery Game Puzzles—7%
DTBench—90.1%
LiveBench Data Analysis64.3%—
LMCA—44.5%
ForecastBench61.7—
LiveBench69%—

Math GPT-6 Luna leads

GPT-4.5: 32.6 (#211), GPT-6 Luna: 76.1 (#15)

BenchmarkGPT-4.5GPT-6 Luna
OTIS Mock AIME 2024-202537.8%98.9%
LMArena Math14121416
FrontierMath (Tiers 1-3)—78.9%
FrontierMath Tier 4—56.1%
ProofBench—64%
LiveBench Math69.3%—
MATH Level 578.6%—

Knowledge GPT-6 Luna leads

GPT-4.5: 32.5 (#211), GPT-6 Luna: 57.0 (#41)

BenchmarkGPT-4.5GPT-6 Luna
GPQA Diamond68.7%90.5%
LMArena Expert13941444
Humanity's Last Exam5.4%—
SimpleQA Verified—41.4%
Confabulations13.6%—

Multimodal GPT-6 Luna leads

GPT-4.5: 37.6 (#71), GPT-6 Luna: 42.4 (#30)

BenchmarkGPT-4.5GPT-6 Luna
LMArena Vision11951217
VPCT45%—
Blueprint-Bench 2—31.2%
Furniture Assembly—44.2%

Multilingual GPT-4.5 leads

GPT-4.5: 52.5 (#83), GPT-6 Luna: 50.5 (#117)

BenchmarkGPT-4.5GPT-6 Luna
LMArena Non-English14131386
LMArena Chinese14211433
LMArena French14181420
LMArena German14571369
LMArena Japanese14161369
LMArena Korean13921360
LMArena Russian14191394
LMArena Spanish—1393

Instruction Following GPT-6 Luna leads

GPT-4.5: 72.6 (#134), GPT-6 Luna: 74.3 (#99)

BenchmarkGPT-4.5GPT-6 Luna
LMArena Instruction Following14041409
LiveBench Instruction Following72.3%—

Long Context GPT-6 Luna leads

GPT-4.5: 40.4 (#155), GPT-6 Luna: 43.0 (#111)

BenchmarkGPT-4.5GPT-6 Luna
LMArena Longer Query14061409
Fiction.LiveBench63.9%—

Writing & Preference GPT-6 Luna leads

GPT-4.5: 56.9 (#134), GPT-6 Luna: 58.3 (#119)

BenchmarkGPT-4.5GPT-6 Luna
LMArena Text14171391
LMArena Creative Writing13941363
LMArena Multi-Turn14441396
Short-Story Creative Writing75.6%—
EQ-Bench Creative Writing1258—
LiveBench Language61.5%—

Frequently asked questions

Is GPT-4.5 better than GPT-6 Luna?

GPT-6 Luna is the stronger model overall, scoring 53.3 to 37.2 on the Noometry Index.

Is GPT-4.5 or GPT-6 Luna better for coding?

GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 42.2 in the Noometry coding category.

How many benchmarks do GPT-4.5 and GPT-6 Luna share?

22 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and GPT-6 Luna has 42.

Related comparisons

Go deeper