Noometry

Model comparison

GPT-5.4 mini vs GPT-6 Luna

GPT-6 Luna is the stronger model overall, scoring 53.3 to 45.0 on the Noometry Index.

Last verified . 37 shared benchmarks.

GPT-5.4 mini OpenAI

45.0

Rank #76 Confirmed

GPT-6 Luna OpenAI

53.3

Rank #36 Confirmed

Summary

  • They share 37 benchmarks with published results for both. GPT-5.4 mini scores higher in 2 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 45.5.
  • The biggest single-benchmark swing is FrontierMath Tier 4: 9.8% for GPT-5.4 mini and 56.1% for GPT-6 Luna.
  • GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $0.75 / $4.50 for GPT-5.4 mini.
  • GPT-6 Luna accepts more context: 1.05M tokens versus 400K.

Side by side

GPT-5.4 miniGPT-6 Luna
ProviderOpenAIOpenAI
Noometry Index45.053.3
Released2026-03-172026-09-22
WeightsProprietaryProprietary
Context window400K1.05M
Max output128K128K
Input $ / M tokens$0.75$0.10
Output $ / M tokens$4.50$0.50
Results tracked4642

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

Category by category

Coding GPT-6 Luna leads

GPT-5.4 mini: 45.2 (#72), GPT-6 Luna: 55.5 (#25)

BenchmarkGPT-5.4 miniGPT-6 Luna
FrontierCode27%42.4%
LMArena WebDev13971581
SciCode49.9%54.6%
LMArena Coding14381439
ALE-Bench1,1891,577
DeepSWE—66.6%
WeirdML60.3%—

Agentic & Tool Use GPT-6 Luna leads

GPT-5.4 mini: 29.9 (#81), GPT-6 Luna: 33.3 (#54)

BenchmarkGPT-5.4 miniGPT-6 Luna
APEX-Agents—44.3%
DeepResearch Bench36.3%—
GDP.pdf—23%

Reasoning GPT-6 Luna leads

GPT-5.4 mini: 30.4 (#85), GPT-6 Luna: 48.2 (#41)

BenchmarkGPT-5.4 miniGPT-6 Luna
ARC-AGI-218.9%59.3%
NYT Connections (extended)61.8%68.7%
ARC-AGI-163.7%86.7%
CritPt10%19.4%
Chess Puzzles24%31%
LMArena Hard Prompts14241411
Mystery Game Puzzles11%7%
DTBench80%90.1%
LMCA40.8%44.5%
Epoch Capabilities Index148.84156.28
Kagi LLM Benchmark37.9%—
Thematic Generalization61.7%—
ForecastBench57—

Math GPT-6 Luna leads

GPT-5.4 mini: 45.5 (#75), GPT-6 Luna: 76.1 (#15)

BenchmarkGPT-5.4 miniGPT-6 Luna
FrontierMath (Tiers 1-3)51.2%78.9%
FrontierMath Tier 49.8%56.1%
OTIS Mock AIME 2024-202588.9%98.9%
ProofBench21%64%
LMArena Math14191416
FrontierMath (Feb 2025 set)28.3%—
FrontierMath Tier 4 (v1)2.1%—

Knowledge GPT-6 Luna leads

GPT-5.4 mini: 51.5 (#67), GPT-6 Luna: 57.0 (#41)

BenchmarkGPT-5.4 miniGPT-6 Luna
GPQA Diamond86.9%90.5%
SimpleQA Verified29.4%41.4%
LMArena Expert14351444
Vectara Hallucination Rate5.5%—

Multimodal GPT-6 Luna leads

GPT-5.4 mini: 39.7 (#56), GPT-6 Luna: 42.4 (#30)

BenchmarkGPT-5.4 miniGPT-6 Luna
LMArena Vision12451217
Blueprint-Bench 2—31.2%
Furniture Assembly—44.2%

Multilingual GPT-5.4 mini leads

GPT-5.4 mini: 51.9 (#96), GPT-6 Luna: 50.5 (#117)

BenchmarkGPT-5.4 miniGPT-6 Luna
LMArena Non-English14051386
LMArena Chinese14461433
LMArena French14401420
LMArena German14091369
LMArena Japanese13741369
LMArena Korean13681360
LMArena Russian14171394
LMArena Spanish14051393

Instruction Following Too close to call

GPT-5.4 mini: 74.1 (#102), GPT-6 Luna: 74.3 (#99)

BenchmarkGPT-5.4 miniGPT-6 Luna
LMArena Instruction Following14051409

Long Context Too close to call

GPT-5.4 mini: 43.0 (#112), GPT-6 Luna: 43.0 (#111)

BenchmarkGPT-5.4 miniGPT-6 Luna
LMArena Longer Query14071409

Writing & Preference GPT-5.4 mini leads

GPT-5.4 mini: 64.0 (#58), GPT-6 Luna: 58.3 (#119)

BenchmarkGPT-5.4 miniGPT-6 Luna
LMArena Text14121391
LMArena Creative Writing13701363
LMArena Multi-Turn14291396
EQ-Bench Creative Writing1665—

Frequently asked questions

Is GPT-5.4 mini better than GPT-6 Luna?

GPT-6 Luna is the stronger model overall, scoring 53.3 to 45.0 on the Noometry Index.

Which is cheaper, GPT-5.4 mini 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-5.4 mini lists at $0.75 and $4.50.

Is GPT-5.4 mini or GPT-6 Luna better for coding?

GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 45.2 in the Noometry coding category.

Which has the bigger context window?

GPT-6 Luna does, with 1.05M tokens against 400K.

How many benchmarks do GPT-5.4 mini and GPT-6 Luna share?

37 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and GPT-6 Luna has 42.

Related comparisons

Go deeper