Noometry

Model comparison

Claude 3.7 Sonnet vs GPT-4.5

Claude 3.7 Sonnet is the stronger model overall, scoring 39.5 to 37.2 on the Noometry Index.

Last verified . 41 shared benchmarks.

Claude 3.7 Sonnet Anthropic

39.5

Rank #164 Confirmed

GPT-4.5 OpenAI

37.2

Rank #208 Confirmed

Summary

  • They share 41 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 6 categories and GPT-4.5 in 4 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in long context, where Claude 3.7 Sonnet leads 50.3 to 40.4.
  • The biggest single-benchmark swing is Aider Polyglot: 64.9% for Claude 3.7 Sonnet and 44.9% for GPT-4.5.

Side by side

Claude 3.7 SonnetGPT-4.5
ProviderAnthropicOpenAI
Noometry Index39.537.2
Released2025-02-242025-02-27
WeightsProprietaryProprietary
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked5842

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

Category by category

Coding GPT-4.5 leads

Claude 3.7 Sonnet: 40.6 (#136), GPT-4.5: 42.2 (#109)

BenchmarkClaude 3.7 SonnetGPT-4.5
Aider Polyglot64.9%44.9%
LiveBench Coding74.5%75.2%
LMArena Coding13611396
SWE-bench Verified61%—
SWE-bench Verified (bash only)52.8%—
GSO3.8%—
WeirdML—39.4%
CadEval54%—

Agentic & Tool Use Claude 3.7 Sonnet leads

Claude 3.7 Sonnet: 34.1 (#50), GPT-4.5: 27.9 (#97)

BenchmarkClaude 3.7 SonnetGPT-4.5
Cybench20%17.5%
TheAgentCompany30.9%—
DeepResearch Bench43.6%—
OSWorld35.8%—
METR Time Horizons60%—

Reasoning Claude 3.7 Sonnet leads

Claude 3.7 Sonnet: 18.6 (#277), GPT-4.5: 13.9 (#330)

BenchmarkClaude 3.7 SonnetGPT-4.5
ARC-AGI-20.9%0.8%
SimpleBench46.4%34.5%
ARC-AGI-128.6%10.3%
EnigmaEval4.2%3.2%
LiveBench Reasoning87.8%71.1%
LMArena Hard Prompts13331403
LiveBench Data Analysis74%64.3%
Epoch Capabilities Index141.16136.74
ForecastBench61.861.7
LiveBench76.1%69%

Math Claude 3.7 Sonnet leads

Claude 3.7 Sonnet: 37.5 (#153), GPT-4.5: 32.6 (#211)

BenchmarkClaude 3.7 SonnetGPT-4.5
OTIS Mock AIME 2024-202557.8%37.8%
LiveBench Math79%69.3%
LMArena Math13371412
MATH Level 591.2%78.6%
Omni-MATH33%—
FrontierMath (Feb 2025 set)4.1%—

Knowledge Claude 3.7 Sonnet leads

Claude 3.7 Sonnet: 39.8 (#130), GPT-4.5: 32.5 (#211)

BenchmarkClaude 3.7 SonnetGPT-4.5
GPQA Diamond79.7%68.7%
Humanity's Last Exam8%5.4%
Confabulations14.7%13.6%
LMArena Expert13211394
MMLU-Pro78.4%—
GPQA (HELM)60.8%—

Multimodal GPT-4.5 leads

Claude 3.7 Sonnet: 33.7 (#95), GPT-4.5: 37.6 (#71)

BenchmarkClaude 3.7 SonnetGPT-4.5
LMArena Vision11691195
VPCT39%45%
GeoBench68%—
SpatialViz-Bench33.9%—

Multilingual GPT-4.5 leads

Claude 3.7 Sonnet: 44.1 (#179), GPT-4.5: 52.5 (#83)

BenchmarkClaude 3.7 SonnetGPT-4.5
LMArena Non-English12961413
LMArena Chinese12991421
LMArena French13031418
LMArena German13011457
LMArena Japanese12671416
LMArena Korean12491392
LMArena Russian13111419
LMArena Spanish1298—

Instruction Following Too close to call

Claude 3.7 Sonnet: 72.9 (#125), GPT-4.5: 72.6 (#134)

BenchmarkClaude 3.7 SonnetGPT-4.5
LiveBench Instruction Following81.3%72.3%
LMArena Instruction Following13521404
IFEval83.4%—

Long Context Claude 3.7 Sonnet leads

Claude 3.7 Sonnet: 50.3 (#10), GPT-4.5: 40.4 (#155)

BenchmarkClaude 3.7 SonnetGPT-4.5
Fiction.LiveBench83.3%63.9%
LMArena Longer Query13731406

Writing & Preference GPT-4.5 leads

Claude 3.7 Sonnet: 54.4 (#150), GPT-4.5: 56.9 (#134)

BenchmarkClaude 3.7 SonnetGPT-4.5
LMArena Text13141417
LMArena Creative Writing13321394
Short-Story Creative Writing81.1%75.6%
EQ-Bench Creative Writing14121258
LMArena Multi-Turn13391444
LiveBench Language59.9%61.5%
WildBench81.4%—

Frequently asked questions

Is Claude 3.7 Sonnet better than GPT-4.5?

Claude 3.7 Sonnet is the stronger model overall, scoring 39.5 to 37.2 on the Noometry Index.

Is Claude 3.7 Sonnet or GPT-4.5 better for coding?

GPT-4.5 scores higher on coding benchmarks: 42.2 versus 40.6 in the Noometry coding category.

How many benchmarks do Claude 3.7 Sonnet and GPT-4.5 share?

41 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and GPT-4.5 has 42.

Related comparisons

Go deeper