Noometry

Model comparison

GPT-6 Luna vs Grok 4

GPT-6 Luna is the stronger model overall, scoring 53.3 to 48.1 on the Noometry Index.

Last verified . 24 shared benchmarks.

GPT-6 Luna OpenAI

53.3

Rank #36 Confirmed

Grok 4 xAI

48.1

Rank #56 Confirmed

Summary

  • They share 24 benchmarks with published results for both. GPT-6 Luna scores higher in 6 categories and Grok 4 in 4 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-6 Luna leads 76.1 to 48.4.
  • The biggest single-benchmark swing is ARC-AGI-2: 59.3% for GPT-6 Luna and 16% for Grok 4.

Side by side

GPT-6 LunaGrok 4
ProviderOpenAIxAI
Noometry Index53.348.1
Released2026-09-222025-07-09
WeightsProprietaryProprietary
Context window1.05M—
Max output128K—
Input $ / M tokens$0.10—
Output $ / M tokens$0.50—
Results tracked4248

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

Category by category

Coding GPT-6 Luna leads

GPT-6 Luna: 55.5 (#25), Grok 4: 50.3 (#46)

BenchmarkGPT-6 LunaGrok 4
LMArena Coding14391408
DeepSWE66.6%—
FrontierCode42.4%—
Aider Polyglot—79.6%
LMArena WebDev1581—
SciCode54.6%—
WeirdML—45.7%
ALE-Bench1,577—

Agentic & Tool Use GPT-6 Luna leads

GPT-6 Luna: 33.3 (#54), Grok 4: 32.3 (#68)

BenchmarkGPT-6 LunaGrok 4
Terminal-Bench—27.2%
APEX-Agents44.3%—
Berkeley Function Calling Leaderboard—63%
GDPval—21.1%
Cybench—43%
DeepResearch Bench—47.3%
BALROG—43.6%
GDP.pdf23%—
LMArena Search—1142
METR Time Horizons—66.6%

Reasoning GPT-6 Luna leads

GPT-6 Luna: 48.2 (#41), Grok 4: 36.7 (#65)

BenchmarkGPT-6 LunaGrok 4
ARC-AGI-259.3%16%
ARC-AGI-186.7%66.7%
Chess Puzzles31%28%
LMArena Hard Prompts14111409
Epoch Capabilities Index156.28146.44
SimpleBench—60.5%
Kagi LLM Benchmark—73.6%
NYT Connections (extended)68.7%—
CritPt19.4%—
Mystery Game Puzzles7%—
DTBench90.1%—
LMCA44.5%—
ForecastBench—60.9

Math GPT-6 Luna leads

GPT-6 Luna: 76.1 (#15), Grok 4: 48.4 (#64)

BenchmarkGPT-6 LunaGrok 4
OTIS Mock AIME 2024-202598.9%84%
LMArena Math14161422
FrontierMath (Tiers 1-3)78.9%—
FrontierMath Tier 456.1%—
ProofBench64%—
Omni-MATH—60.3%
FrontierMath (Feb 2025 set)—19.7%
FrontierMath Tier 4 (v1)—2.1%

Knowledge GPT-6 Luna leads

GPT-6 Luna: 57.0 (#41), Grok 4: 53.8 (#55)

BenchmarkGPT-6 LunaGrok 4
GPQA Diamond90.5%87%
LMArena Expert14441415
SimpleQA Verified41.4%—
MMLU-Pro—85.1%
Confabulations—12.4%
GPQA (HELM)—72.7%

Multimodal GPT-6 Luna leads

GPT-6 Luna: 42.4 (#30), Grok 4: 33.7 (#94)

BenchmarkGPT-6 LunaGrok 4
LMArena Vision12171210
GeoBench—45%
Blueprint-Bench 231.2%—
Furniture Assembly44.2%—

Multilingual Grok 4 leads

GPT-6 Luna: 50.5 (#117), Grok 4: 51.8 (#103)

BenchmarkGPT-6 LunaGrok 4
LMArena Non-English13861403
LMArena Chinese14331427
LMArena French14201418
LMArena German13691429
LMArena Japanese13691394
LMArena Korean13601377
LMArena Russian13941410
LMArena Spanish13931420

Instruction Following Grok 4 leads

GPT-6 Luna: 74.3 (#99), Grok 4: 79.2 (#5)

BenchmarkGPT-6 LunaGrok 4
LMArena Instruction Following14091387
IFEval—94.9%

Long Context Grok 4 leads

GPT-6 Luna: 43.0 (#111), Grok 4: 63.1 (#4)

BenchmarkGPT-6 LunaGrok 4
LMArena Longer Query14091409
Fiction.LiveBench—94.4%

Writing & Preference Too close to call

GPT-6 Luna: 58.3 (#119), Grok 4: 58.5 (#116)

BenchmarkGPT-6 LunaGrok 4
LMArena Text13911411
LMArena Creative Writing13631397
LMArena Multi-Turn13961416
Short-Story Creative Writing—76.9%
WildBench—79.7%

Frequently asked questions

Is GPT-6 Luna better than Grok 4?

GPT-6 Luna is the stronger model overall, scoring 53.3 to 48.1 on the Noometry Index.

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

GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 50.3 in the Noometry coding category.

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

24 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and Grok 4 has 48.

Related comparisons

Go deeper