Noometry

Model comparison

Claude Sonnet 4 vs GPT-5.5

GPT-5.5 is the stronger model overall, scoring 63.4 to 40.6 on the Noometry Index. Claude Sonnet 4 costs 1.9× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

Last verified . 38 shared benchmarks.

Claude Sonnet 4 Anthropic

40.6

Rank #139 Confirmed

GPT-5.5 OpenAI

63.4

Rank #8 Confirmed

Summary

  • They share 38 benchmarks with published results for both. Claude Sonnet 4 scores higher in 0 categories and GPT-5.5 in 10 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where GPT-5.5 leads 72.7 to 23.0.
  • The biggest single-benchmark swing is ARC-AGI-2: 5.9% for Claude Sonnet 4 and 85% for GPT-5.5.
  • Claude Sonnet 4 is cheaper at $3 / $15 per million input/output tokens, against $5 / $30 for GPT-5.5.
  • GPT-5.5 accepts more context: 1.05M tokens versus 200K.

Side by side

Claude Sonnet 4GPT-5.5
ProviderAnthropicOpenAI
Noometry Index40.663.4
Released2025-05-222026-04-23
WeightsProprietaryProprietary
Context window200K1.05M
Max output64K128K
Input $ / M tokens$3$5
Output $ / M tokens$15$30
Results tracked5871

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

Category by category

Coding GPT-5.5 leads

Claude Sonnet 4: 44.0 (#70), GPT-5.5: 58.3 (#13)

BenchmarkClaude Sonnet 4GPT-5.5
SciCode40%56.1%
GSO4.9%40.2%
WeirdML46.1%84.9%
LMArena Coding14141494
ALE-Bench655.351,943
SWE-bench Verified—80.6%
DeepSWE—67%
FrontierCode—43%
SWE-bench Verified (bash only)64.9%—
Aider Polyglot61.3%—
LMArena WebDev—1512
MirrorCode—10%

Agentic & Tool Use GPT-5.5 leads

Claude Sonnet 4: 38.3 (#29), GPT-5.5: 50.7 (#6)

BenchmarkClaude Sonnet 4GPT-5.5
DeepResearch Bench46.6%54%
Terminal-Bench—84.7%
APEX-Agents—55.1%
OSWorld 2.0—13%
Remote Labor Index—6.3%
TheAgentCompany33.1%—
τ²-bench Banking—44.6%
Cybench35%—
OSWorld43.9%—
PostTrainBench—27.2%
ExploitBench—47.4%
GBAEval—53.2%
GDP.pdf—26%
LMArena Search—1242
METR Time Horizons62%—
Vending-Bench 2—7,524

Reasoning GPT-5.5 leads

Claude Sonnet 4: 23.0 (#169), GPT-5.5: 72.7 (#8)

BenchmarkClaude Sonnet 4GPT-5.5
ARC-AGI-25.9%85%
SimpleBench45.5%69%
Kagi LLM Benchmark73%88.8%
ARC-AGI-140%95%
CritPt0.3%27.1%
LMArena Hard Prompts13721489
DTBench77.1%96%
LMCA29%54.3%
Epoch Capabilities Index141.69159.25
ForecastBench60.260.6
NYT Connections (extended)—96.2%
Chess Puzzles—54%
EnigmaEval3.1%—
EBR-Bench—34.3%
Mystery Game Puzzles—56%
Surface Evolver Bench—88.1%
Bench to the Future 3—0.14

Math GPT-5.5 leads

Claude Sonnet 4: 41.5 (#71), GPT-5.5: 81.5 (#9)

BenchmarkClaude Sonnet 4GPT-5.5
OTIS Mock AIME 2024-202571.1%100%
LMArena Math13751486
FrontierMath (Feb 2025 set)4.1%51.7%
FrontierMath Tier 4 (v1)0%35.4%
FrontierMath (Tiers 1-3)—85.3%
FrontierMath Tier 4—72.5%
MathArena Final-Answer Competitions—94.3%
ProofBench—50%
Omni-MATH60.2%—
MATH Level 584.4%—
FrontierMath Erdős—0%

Knowledge GPT-5.5 leads

Claude Sonnet 4: 41.6 (#99), GPT-5.5: 64.4 (#14)

BenchmarkClaude Sonnet 4GPT-5.5
GPQA Diamond79.2%94%
Vectara Hallucination Rate10.3%9.3%
LMArena Expert13711508
Humanity's Last Exam7.8%—
SimpleQA Verified—63%
MMLU-Pro84.3%—
Confabulations13.2%—
GPQA (HELM)70.6%—

Multimodal GPT-5.5 leads

Claude Sonnet 4: 25.6 (#116), GPT-5.5: 47.1 (#11)

BenchmarkClaude Sonnet 4GPT-5.5
LMArena Vision11911297
GeoBench37%—
VPCT34%—
Blueprint-Bench 2—36.2%
Furniture Assembly—44.2%
LMArena Document—1486
MindCube44.8%—

Multilingual GPT-5.5 leads

Claude Sonnet 4: 46.9 (#151), GPT-5.5: 56.5 (#19)

BenchmarkClaude Sonnet 4GPT-5.5
LMArena Non-English13331467
LMArena Chinese13511535
LMArena French13641486
LMArena German13311478
LMArena Japanese13011498
LMArena Korean12931461
LMArena Russian13551473
LMArena Spanish13561469

Instruction Following GPT-5.5 leads

Claude Sonnet 4: 71.8 (#141), GPT-5.5: 77.7 (#18)

BenchmarkClaude Sonnet 4GPT-5.5
LMArena Instruction Following13751479
IFEval84%—

Long Context GPT-5.5 leads

Claude Sonnet 4: 33.6 (#254), GPT-5.5: 48.3 (#12)

BenchmarkClaude Sonnet 4GPT-5.5
LMArena Longer Query13981484
Fiction.LiveBench46.9%—
CL-bench Life—22.2%

Writing & Preference GPT-5.5 leads

Claude Sonnet 4: 57.3 (#127), GPT-5.5: 72.8 (#13)

BenchmarkClaude Sonnet 4GPT-5.5
LMArena Text13511471
LMArena Creative Writing13451455
EQ-Bench Creative Writing14831844
LMArena Multi-Turn13761477
Short-Story Creative Writing81.4%—
WildBench83.8%—
EQ-Bench 4—1315

Frequently asked questions

Is Claude Sonnet 4 better than GPT-5.5?

GPT-5.5 is the stronger model overall, scoring 63.4 to 40.6 on the Noometry Index. Claude Sonnet 4 costs 1.9× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

Which is cheaper, Claude Sonnet 4 or GPT-5.5?

Claude Sonnet 4 is cheaper. It lists at $3 per million input tokens and $15 per million output tokens; GPT-5.5 lists at $5 and $30.

Is Claude Sonnet 4 or GPT-5.5 better for coding?

GPT-5.5 scores higher on coding benchmarks: 58.3 versus 44.0 in the Noometry coding category.

Which has the bigger context window?

GPT-5.5 does, with 1.05M tokens against 200K.

How many benchmarks do Claude Sonnet 4 and GPT-5.5 share?

38 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and GPT-5.5 has 71.

Related comparisons

Go deeper