Noometry

Model comparison

GPT-4 vs GPT-6 Luna

GPT-6 Luna is the stronger model overall, scoring 53.3 to 29.1 on the Noometry Index.

Last verified . 24 shared benchmarks.

GPT-4 OpenAI

29.1

Rank #316 Confirmed

GPT-6 Luna OpenAI

53.3

Rank #36 Confirmed

Summary

  • They share 24 benchmarks with published results for both. GPT-4 scores higher in 0 categories and GPT-6 Luna in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-6 Luna leads 76.1 to 10.8.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.1% for GPT-4 and 98.9% for GPT-6 Luna.
  • GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $30 / $60 for GPT-4.
  • GPT-6 Luna accepts more context: 1.05M tokens versus 8K.

Side by side

GPT-4GPT-6 Luna
ProviderOpenAIOpenAI
Noometry Index29.153.3
Released2023-03-142026-09-22
WeightsProprietaryProprietary
Context window8K1.05M
Max output8K128K
Input $ / M tokens$30$0.10
Output $ / M tokens$60$0.50
Results tracked3842

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

Category by category

Coding GPT-6 Luna leads

GPT-4: 31.6 (#283), GPT-6 Luna: 55.5 (#25)

BenchmarkGPT-4GPT-6 Luna
LMArena Coding12541439
DeepSWE—66.6%
FrontierCode—42.4%
LMArena WebDev—1581
SciCode—54.6%
WeirdML12.4%—
BigCodeBench Instruct46%—
BigCodeBench Complete57.2%—
ALE-Bench—1,577
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, GPT-6 Luna: 33.3 (#54)

BenchmarkGPT-4GPT-6 Luna
APEX-Agents—44.3%
GDP.pdf—23%
METR Time Horizons36.1%—

Reasoning GPT-6 Luna leads

GPT-4: 17.8 (#289), GPT-6 Luna: 48.2 (#41)

BenchmarkGPT-4GPT-6 Luna
Chess Puzzles4%31%
LMArena Hard Prompts12411411
Mystery Game Puzzles12%7%
DTBench62.7%90.1%
LMCA17.1%44.5%
Epoch Capabilities Index125.89156.28
ARC-AGI-2—59.3%
NYT Connections (extended)—68.7%
ARC-AGI-1—86.7%
CritPt—19.4%
BIG-Bench Hard75.1%—
ForecastBench57.8—
HellaSwag95.3%—
WinoGrande87.5%—

Math GPT-6 Luna leads

GPT-4: 10.8 (#309), GPT-6 Luna: 76.1 (#15)

BenchmarkGPT-4GPT-6 Luna
OTIS Mock AIME 2024-20251.1%98.9%
LMArena Math12691416
FrontierMath (Tiers 1-3)—78.9%
FrontierMath Tier 4—56.1%
ProofBench—64%
MATH Level 523%—
GSM8K92%—

Knowledge GPT-6 Luna leads

GPT-4: 18.4 (#282), GPT-6 Luna: 57.0 (#41)

BenchmarkGPT-4GPT-6 Luna
GPQA Diamond35.7%90.5%
LMArena Expert12111444
SimpleQA Verified—41.4%
MMLU86.4%—
TriviaQA84.8%—

Multimodal Not comparable

GPT-4: —, GPT-6 Luna: 42.4 (#30)

BenchmarkGPT-4GPT-6 Luna
LMArena Vision—1217
Blueprint-Bench 2—31.2%
Furniture Assembly—44.2%

Multilingual GPT-6 Luna leads

GPT-4: 40.6 (#215), GPT-6 Luna: 50.5 (#117)

BenchmarkGPT-4GPT-6 Luna
LMArena Non-English12461386
LMArena Chinese12421433
LMArena French12831420
LMArena German12511369
LMArena Japanese12091369
LMArena Korean11841360
LMArena Russian12511394
LMArena Spanish12611393

Instruction Following GPT-6 Luna leads

GPT-4: 65.3 (#222), GPT-6 Luna: 74.3 (#99)

BenchmarkGPT-4GPT-6 Luna
LMArena Instruction Following12411409

Long Context GPT-6 Luna leads

GPT-4: 37.7 (#212), GPT-6 Luna: 43.0 (#111)

BenchmarkGPT-4GPT-6 Luna
LMArena Longer Query12441409

Writing & Preference GPT-6 Luna leads

GPT-4: 34.9 (#268), GPT-6 Luna: 58.3 (#119)

BenchmarkGPT-4GPT-6 Luna
LMArena Text12631391
LMArena Creative Writing12441363
LMArena Multi-Turn12571396
EQ-Bench Creative Writing752—

Frequently asked questions

Is GPT-4 better than GPT-6 Luna?

GPT-6 Luna is the stronger model overall, scoring 53.3 to 29.1 on the Noometry Index.

Which is cheaper, GPT-4 or GPT-6 Luna?

GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; GPT-4 lists at $30 and $60.

Is GPT-4 or GPT-6 Luna better for coding?

GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 31.6 in the Noometry coding category.

Which has the bigger context window?

GPT-6 Luna does, with 1.05M tokens against 8K.

How many benchmarks do GPT-4 and GPT-6 Luna share?

24 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and GPT-6 Luna has 42.

Related comparisons

Go deeper