Noometry

Model comparison

Claude Sonnet 4.5 vs GPT-5.6 Luna

GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 44.1 on the Noometry Index.

Last verified . 41 shared benchmarks.

Claude Sonnet 4.5 Anthropic

44.1

Rank #81 Confirmed

GPT-5.6 Luna OpenAI

54.6

Rank #30 Confirmed

Summary

  • They share 41 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 3 categories and GPT-5.6 Luna in 7 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-5.6 Luna leads 77.7 to 32.3.
  • The biggest single-benchmark swing is FrontierMath Tier 4: 2.4% for Claude Sonnet 4.5 and 61% for GPT-5.6 Luna.
  • GPT-5.6 Luna is cheaper at $0.20 / $1.20 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.5.
  • GPT-5.6 Luna accepts more context: 1.05M tokens versus 200K.

Side by side

Claude Sonnet 4.5GPT-5.6 Luna
ProviderAnthropicOpenAI
Noometry Index44.154.6
Released2025-09-292026-07-09
WeightsProprietaryProprietary
Context window200K1.05M
Max output64K128K
Input $ / M tokens$3$0.20
Output $ / M tokens$15$1.20
Results tracked7352

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

Category by category

Coding GPT-5.6 Luna leads

Claude Sonnet 4.5: 47.3 (#61), GPT-5.6 Luna: 54.5 (#28)

BenchmarkClaude Sonnet 4.5GPT-5.6 Luna
LMArena WebDev13931519
SciCode44.7%53.6%
WeirdML47.7%60.9%
LMArena Coding14891466
ALE-Bench796.151,667
SWE-bench Verified71.3%—
DeepSWE—67.2%
FrontierCode—39.8%
SWE-bench Verified (bash only)71.4%—
CursorBench—35.9%
SWE-bench Multilingual67%—
GSO14.7%—
AlgoTune1.52—

Agentic & Tool Use Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 38.3 (#32), GPT-5.6 Luna: 34.4 (#45)

BenchmarkClaude Sonnet 4.5GPT-5.6 Luna
Vending-Bench 23,8394,095
Terminal-Bench46.5%—
APEX-Agents—43%
Berkeley Function Calling Leaderboard73.2%—
GDPval42.5%—
Remote Labor Index2.1%—
τ²-bench Airline72%—
τ²-bench Banking25.3%—
τ²-bench Retail72.4%—
τ²-bench Telecom84.9%—
Cybench60%—
DeepResearch Bench52.6%—
OSWorld62.9%—
BALROG—45.6%
GDP.pdf—22.7%
LMArena Search1159—
METR Time Horizons67.4%—

Reasoning GPT-5.6 Luna leads

Claude Sonnet 4.5: 26.9 (#125), GPT-5.6 Luna: 47.6 (#43)

BenchmarkClaude Sonnet 4.5GPT-5.6 Luna
ARC-AGI-213.6%59.5%
SimpleBench54.3%46.8%
Kagi LLM Benchmark57.9%49.1%
NYT Connections (extended)37.3%69.4%
ARC-AGI-163.7%88%
CritPt1.1%20.6%
Chess Puzzles12%40%
LMArena Hard Prompts14621451
Mystery Game Puzzles17%21%
DTBench83.2%89.1%
LMCA38.8%48.5%
Epoch Capabilities Index146.84156.39
EnigmaEval6%—
EBR-Bench2.4%—
Surface Evolver Bench—61.9%
ForecastBench61.9—

Math GPT-5.6 Luna leads

Claude Sonnet 4.5: 32.3 (#216), GPT-5.6 Luna: 77.7 (#14)

BenchmarkClaude Sonnet 4.5GPT-5.6 Luna
FrontierMath (Tiers 1-3)23.9%82.1%
FrontierMath Tier 42.4%61%
OTIS Mock AIME 2024-202577.8%98.3%
ProofBench19%60%
LMArena Math14491458
Omni-MATH55.3%—
MATH Level 597.7%—
FrontierMath (Feb 2025 set)15.2%—
FrontierMath Tier 4 (v1)4.2%—

Knowledge GPT-5.6 Luna leads

Claude Sonnet 4.5: 48.4 (#76), GPT-5.6 Luna: 58.5 (#34)

BenchmarkClaude Sonnet 4.5GPT-5.6 Luna
GPQA Diamond82.3%91.6%
SimpleQA Verified30.7%41%
LMArena Expert14821478
Humanity's Last Exam13.7%—
MMLU-Pro86.9%—
Vectara Hallucination Rate12%—
GPQA (HELM)68.6%—

Multimodal GPT-5.6 Luna leads

Claude Sonnet 4.5: 34.8 (#89), GPT-5.6 Luna: 42.7 (#28)

BenchmarkClaude Sonnet 4.5GPT-5.6 Luna
LMArena Document14501457
LMArena Vision—1258
VPCT39.8%—
Blueprint-Bench 2—22.6%
Furniture Assembly—42.5%

Multilingual Too close to call

Claude Sonnet 4.5: 53.4 (#69), GPT-5.6 Luna: 52.8 (#78)

BenchmarkClaude Sonnet 4.5GPT-5.6 Luna
LMArena Non-English14251417
LMArena Chinese14591470
LMArena French14581456
LMArena German14271454
LMArena Japanese13901411
LMArena Korean14031415
LMArena Russian14371428
LMArena Spanish14571448

Instruction Following Too close to call

Claude Sonnet 4.5: 75.0 (#78), GPT-5.6 Luna: 75.6 (#57)

BenchmarkClaude Sonnet 4.5GPT-5.6 Luna
LMArena Instruction Following14591437
IFEval85%—

Long Context Claude Sonnet 4.5 leads

Claude Sonnet 4.5: 45.2 (#46), GPT-5.6 Luna: 43.9 (#82)

BenchmarkClaude Sonnet 4.5GPT-5.6 Luna
LMArena Longer Query14761436

Writing & Preference GPT-5.6 Luna leads

Claude Sonnet 4.5: 66.5 (#34), GPT-5.6 Luna: 68.0 (#29)

BenchmarkClaude Sonnet 4.5GPT-5.6 Luna
LMArena Text14391431
LMArena Creative Writing14421396
EQ-Bench Creative Writing16781829
LMArena Multi-Turn14651434
WildBench85.4%—
EQ-Bench 4—1156

Frequently asked questions

Is Claude Sonnet 4.5 better than GPT-5.6 Luna?

GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 44.1 on the Noometry Index.

Which is cheaper, Claude Sonnet 4.5 or GPT-5.6 Luna?

GPT-5.6 Luna is cheaper. It lists at $0.20 per million input tokens and $1.20 per million output tokens; Claude Sonnet 4.5 lists at $3 and $15.

Is Claude Sonnet 4.5 or GPT-5.6 Luna better for coding?

GPT-5.6 Luna scores higher on coding benchmarks: 54.5 versus 47.3 in the Noometry coding category.

Which has the bigger context window?

GPT-5.6 Luna does, with 1.05M tokens against 200K.

How many benchmarks do Claude Sonnet 4.5 and GPT-5.6 Luna share?

41 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and GPT-5.6 Luna has 52.

Related comparisons

Go deeper