Noometry

Model comparison

Claude 3.5 Haiku vs GPT-4.5

GPT-4.5 is the stronger model overall, scoring 36.9 to 28.9 on the Noometry Index.

Last verified . 33 shared benchmarks.

Claude 3.5 Haiku Anthropic

28.9

Rank #309 Confirmed

GPT-4.5 OpenAI

36.9

Rank #200 Confirmed

Summary

  • They share 33 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 2 categories and GPT-4.5 in 8 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-4.5 leads 31.0 to 13.7.
  • The biggest single-benchmark swing is LiveBench Reasoning: 28.1% for Claude 3.5 Haiku and 71.1% for GPT-4.5.

Side by side

Claude 3.5 HaikuGPT-4.5
ProviderAnthropicOpenAI
Noometry Index28.936.9
Released2024-10-222025-02-27
WeightsProprietaryProprietary
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked4942

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

Category by category

Coding GPT-4.5 leads

Claude 3.5 Haiku: 32.6 (#258), GPT-4.5: 41.8 (#102)

BenchmarkClaude 3.5 HaikuGPT-4.5
Aider Polyglot28%44.9%
WeirdML30.7%39.4%
LiveBench Coding51.4%75.2%
LMArena Coding12871397
SciCode27.4%—
BigCodeBench Instruct46.1%—
BigCodeBench Complete59%—
CadEval32%—

Agentic & Tool Use Too close to call

Claude 3.5 Haiku: 27.7 (#90), GPT-4.5: 27.6 (#92)

BenchmarkClaude 3.5 HaikuGPT-4.5
Cybench—17.5%
BALROG19.3%—

Reasoning Claude 3.5 Haiku leads

Claude 3.5 Haiku: 17.1 (#286), GPT-4.5: 14.1 (#320)

BenchmarkClaude 3.5 HaikuGPT-4.5
LiveBench Reasoning28.1%71.1%
LMArena Hard Prompts12511404
LiveBench Data Analysis48.5%64.3%
Epoch Capabilities Index127.15136.74
LiveBench43.5%69%
ARC-AGI-2—0.8%
SimpleBench—34.5%
ARC-AGI-1—10.3%
CritPt0%—
EnigmaEval—3.2%
DTBench56.7%—
ForecastBench—61.7

Math GPT-4.5 leads

Claude 3.5 Haiku: 13.7 (#294), GPT-4.5: 31.0 (#200)

BenchmarkClaude 3.5 HaikuGPT-4.5
OTIS Mock AIME 2024-20254.3%37.8%
LiveBench Math35.5%69.3%
LMArena Math12441412
MATH Level 546.4%78.6%
Omni-MATH22.4%—
FrontierMath (Feb 2025 set)0.3%—

Knowledge GPT-4.5 leads

Claude 3.5 Haiku: 18.7 (#276), GPT-4.5: 32.4 (#205)

BenchmarkClaude 3.5 HaikuGPT-4.5
GPQA Diamond38.1%68.7%
Confabulations36.7%13.6%
LMArena Expert12081394
Humanity's Last Exam—5.4%
MMLU-Pro60.5%—
GPQA (HELM)36.3%—
MMLU74.3%—

Multimodal GPT-4.5 leads

Claude 3.5 Haiku: 26.3 (#113), GPT-4.5: 37.0 (#68)

BenchmarkClaude 3.5 HaikuGPT-4.5
LMArena Vision10921195
GeoBench34%—
VPCT—45%

Multilingual GPT-4.5 leads

Claude 3.5 Haiku: 40.2 (#213), GPT-4.5: 52.7 (#78)

BenchmarkClaude 3.5 HaikuGPT-4.5
LMArena Non-English12381413
LMArena Chinese12291421
LMArena French12621417
LMArena German12371457
LMArena Japanese11751416
LMArena Korean11721392
LMArena Russian12531419
LMArena Spanish1261—

Instruction Following GPT-4.5 leads

Claude 3.5 Haiku: 63.1 (#229), GPT-4.5: 72.8 (#129)

BenchmarkClaude 3.5 HaikuGPT-4.5
LiveBench Instruction Following61.9%72.3%
LMArena Instruction Following12411404
IFEval79.2%—

Long Context GPT-4.5 leads

Claude 3.5 Haiku: 38.3 (#195), GPT-4.5: 40.4 (#150)

BenchmarkClaude 3.5 HaikuGPT-4.5
LMArena Longer Query12611406
Fiction.LiveBench—63.9%

Writing & Preference GPT-4.5 leads

Claude 3.5 Haiku: 42.9 (#228), GPT-4.5: 57.1 (#129)

BenchmarkClaude 3.5 HaikuGPT-4.5
LMArena Text12551417
LMArena Creative Writing12331394
Short-Story Creative Writing73.5%75.6%
EQ-Bench Creative Writing11461258
LMArena Multi-Turn12651444
LiveBench Language35.4%61.5%
WildBench76%—

Frequently asked questions

Is Claude 3.5 Haiku better than GPT-4.5?

GPT-4.5 is the stronger model overall, scoring 36.9 to 28.9 on the Noometry Index.

Is Claude 3.5 Haiku or GPT-4.5 better for coding?

GPT-4.5 scores higher on coding benchmarks: 41.8 versus 32.6 in the Noometry coding category.

How many benchmarks do Claude 3.5 Haiku and GPT-4.5 share?

33 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and GPT-4.5 has 42.

Related comparisons

Go deeper