Noometry

Model comparison

GPT-4 Turbo vs GPT-6 Luna

GPT-6 Luna is the stronger model overall, scoring 53.3 to 30.5 on the Noometry Index.

Last verified . 25 shared benchmarks.

GPT-4 Turbo OpenAI

30.5

Rank #292 Confirmed

GPT-6 Luna OpenAI

53.3

Rank #36 Confirmed

Summary

  • They share 25 benchmarks with published results for both. GPT-4 Turbo scores higher in 0 categories and GPT-6 Luna in 9 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-6 Luna leads 76.1 to 9.0.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.7% for GPT-4 Turbo and 98.9% for GPT-6 Luna.
  • GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
  • GPT-6 Luna accepts more context: 1.05M tokens versus 128K.

Side by side

GPT-4 TurboGPT-6 Luna
ProviderOpenAIOpenAI
Noometry Index30.553.3
Released2023-11-062026-09-22
WeightsProprietaryProprietary
Context window128K1.05M
Max output4K128K
Input $ / M tokens$10$0.10
Output $ / M tokens$30$0.50
Results tracked3642

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

Category by category

Coding GPT-6 Luna leads

GPT-4 Turbo: 33.8 (#249), GPT-6 Luna: 55.5 (#25)

BenchmarkGPT-4 TurboGPT-6 Luna
LMArena Coding12681439
DeepSWE—66.6%
FrontierCode—42.4%
LMArena WebDev—1581
SciCode—54.6%
WeirdML18%—
BigCodeBench Instruct48.2%—
BigCodeBench Complete58.2%—
ALE-Bench—1,577
HumanEval+86.6%—
MBPP+73.3%—

Agentic & Tool Use Not comparable

GPT-4 Turbo: —, GPT-6 Luna: 33.3 (#54)

BenchmarkGPT-4 TurboGPT-6 Luna
APEX-Agents—44.3%
GDP.pdf—23%
METR Time Horizons36.7%—

Reasoning GPT-6 Luna leads

GPT-4 Turbo: 15.3 (#317), GPT-6 Luna: 48.2 (#41)

BenchmarkGPT-4 TurboGPT-6 Luna
Chess Puzzles6%31%
LMArena Hard Prompts12511411
DTBench61.6%90.1%
LMCA9.8%44.5%
Epoch Capabilities Index127.25156.28
ARC-AGI-2—59.3%
SimpleBench25.1%—
NYT Connections (extended)—68.7%
ARC-AGI-1—86.7%
CritPt—19.4%
Mystery Game Puzzles—7%
ForecastBench59.4—

Math GPT-6 Luna leads

GPT-4 Turbo: 9.0 (#322), GPT-6 Luna: 76.1 (#15)

BenchmarkGPT-4 TurboGPT-6 Luna
FrontierMath (Tiers 1-3)0.7%78.9%
OTIS Mock AIME 2024-20256.7%98.9%
LMArena Math12721416
FrontierMath Tier 4—56.1%
ProofBench—64%
MATH Level 546.7%—

Knowledge GPT-6 Luna leads

GPT-4 Turbo: 24.3 (#268), GPT-6 Luna: 57.0 (#41)

BenchmarkGPT-4 TurboGPT-6 Luna
GPQA Diamond46.6%90.5%
LMArena Expert12231444
SimpleQA Verified—41.4%
Confabulations28.4%—
MMLU81.3%—

Multimodal GPT-6 Luna leads

GPT-4 Turbo: 30.6 (#110), GPT-6 Luna: 42.4 (#30)

BenchmarkGPT-4 TurboGPT-6 Luna
LMArena Vision10901217
Blueprint-Bench 2—31.2%
Furniture Assembly—44.2%

Multilingual GPT-6 Luna leads

GPT-4 Turbo: 40.5 (#216), GPT-6 Luna: 50.5 (#117)

BenchmarkGPT-4 TurboGPT-6 Luna
LMArena Non-English12451386
LMArena Chinese12421433
LMArena French12761420
LMArena German12591369
LMArena Japanese11941369
LMArena Korean11871360
LMArena Russian12591394
LMArena Spanish12601393

Instruction Following GPT-6 Luna leads

GPT-4 Turbo: 65.8 (#216), GPT-6 Luna: 74.3 (#99)

BenchmarkGPT-4 TurboGPT-6 Luna
LMArena Instruction Following12491409

Long Context GPT-6 Luna leads

GPT-4 Turbo: 38.0 (#206), GPT-6 Luna: 43.0 (#111)

BenchmarkGPT-4 TurboGPT-6 Luna
LMArena Longer Query12541409

Writing & Preference GPT-6 Luna leads

GPT-4 Turbo: 47.7 (#206), GPT-6 Luna: 58.3 (#119)

BenchmarkGPT-4 TurboGPT-6 Luna
LMArena Text12721391
LMArena Creative Writing12691363
LMArena Multi-Turn12671396

Frequently asked questions

Is GPT-4 Turbo better than GPT-6 Luna?

GPT-6 Luna is the stronger model overall, scoring 53.3 to 30.5 on the Noometry Index.

Which is cheaper, GPT-4 Turbo 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 Turbo lists at $10 and $30.

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

GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 33.8 in the Noometry coding category.

Which has the bigger context window?

GPT-6 Luna does, with 1.05M tokens against 128K.

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

25 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and GPT-6 Luna has 42.

Related comparisons

Go deeper