Noometry

Model comparison

Claude 3 Opus vs GPT-4.5

GPT-4.5 is the stronger model overall, scoring 37.2 to 29.5 on the Noometry Index.

Last verified . 34 shared benchmarks.

Claude 3 Opus Anthropic

29.5

Rank #310 Confirmed

GPT-4.5 OpenAI

37.2

Rank #208 Confirmed

Summary

  • They share 34 benchmarks with published results for both. Claude 3 Opus scores higher in 1 category and GPT-4.5 in 9 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-4.5 leads 32.6 to 14.8.
  • The biggest single-benchmark swing is MATH Level 5: 37.5% for Claude 3 Opus and 78.6% for GPT-4.5.

Side by side

Claude 3 OpusGPT-4.5
ProviderAnthropicOpenAI
Noometry Index29.537.2
Released2024-02-292025-02-27
WeightsProprietaryProprietary
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked4642

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

Category by category

Coding GPT-4.5 leads

Claude 3 Opus: 32.9 (#267), GPT-4.5: 42.2 (#109)

BenchmarkClaude 3 OpusGPT-4.5
WeirdML19.2%39.4%
LiveBench Coding38.6%75.2%
LMArena Coding12641396
Aider Polyglot—44.9%
BigCodeBench Instruct45.5%—
BigCodeBench Complete57.4%—
HumanEval+77.4%—
MBPP+73.3%—

Agentic & Tool Use GPT-4.5 leads

Claude 3 Opus: 24.6 (#116), GPT-4.5: 27.9 (#97)

BenchmarkClaude 3 OpusGPT-4.5
Cybench10%17.5%
METR Time Horizons29.5%—

Reasoning Too close to call

Claude 3 Opus: 14.6 (#324), GPT-4.5: 13.9 (#330)

BenchmarkClaude 3 OpusGPT-4.5
SimpleBench23.5%34.5%
EnigmaEval0.8%3.2%
LiveBench Reasoning40.6%71.1%
LMArena Hard Prompts12451403
LiveBench Data Analysis57.9%64.3%
Epoch Capabilities Index126.91136.74
ForecastBench58.461.7
LiveBench49.2%69%
ARC-AGI-2—0.8%
ARC-AGI-1—10.3%
Chess Puzzles5%—
DTBench61.6%—
LMCA17%—
WinoGrande88.5%—

Math GPT-4.5 leads

Claude 3 Opus: 14.8 (#299), GPT-4.5: 32.6 (#211)

BenchmarkClaude 3 OpusGPT-4.5
OTIS Mock AIME 2024-20254.7%37.8%
LiveBench Math43.6%69.3%
LMArena Math12731412
MATH Level 537.5%78.6%

Knowledge GPT-4.5 leads

Claude 3 Opus: 24.5 (#267), GPT-4.5: 32.5 (#211)

BenchmarkClaude 3 OpusGPT-4.5
GPQA Diamond47.2%68.7%
Confabulations22.7%13.6%
LMArena Expert12231394
Humanity's Last Exam—5.4%
SimpleQA Verified12.6%—
MMLU84.6%—

Multimodal GPT-4.5 leads

Claude 3 Opus: 27.1 (#116), GPT-4.5: 37.6 (#71)

BenchmarkClaude 3 OpusGPT-4.5
LMArena Vision10231195
VPCT—45%

Multilingual GPT-4.5 leads

Claude 3 Opus: 41.4 (#207), GPT-4.5: 52.5 (#83)

BenchmarkClaude 3 OpusGPT-4.5
LMArena Non-English12581413
LMArena Chinese12481421
LMArena French12751418
LMArena German12581457
LMArena Japanese12041416
LMArena Korean11871392
LMArena Russian12801419
LMArena Spanish1246—

Instruction Following GPT-4.5 leads

Claude 3 Opus: 64.1 (#228), GPT-4.5: 72.6 (#134)

BenchmarkClaude 3 OpusGPT-4.5
LiveBench Instruction Following63.9%72.3%
LMArena Instruction Following12481404

Long Context GPT-4.5 leads

Claude 3 Opus: 38.2 (#202), GPT-4.5: 40.4 (#155)

BenchmarkClaude 3 OpusGPT-4.5
LMArena Longer Query12591406
Fiction.LiveBench—63.9%

Writing & Preference GPT-4.5 leads

Claude 3 Opus: 47.2 (#213), GPT-4.5: 56.9 (#134)

BenchmarkClaude 3 OpusGPT-4.5
LMArena Text12621417
LMArena Creative Writing12351394
LMArena Multi-Turn12751444
LiveBench Language50.4%61.5%
Short-Story Creative Writing—75.6%
EQ-Bench Creative Writing—1258

Frequently asked questions

Is Claude 3 Opus better than GPT-4.5?

GPT-4.5 is the stronger model overall, scoring 37.2 to 29.5 on the Noometry Index.

Is Claude 3 Opus or GPT-4.5 better for coding?

GPT-4.5 scores higher on coding benchmarks: 42.2 versus 32.9 in the Noometry coding category.

How many benchmarks do Claude 3 Opus and GPT-4.5 share?

34 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and GPT-4.5 has 42.

Related comparisons

Go deeper