Noometry

Model comparison

GPT-4.5 vs o1

o1 is the stronger model overall, scoring 40.9 to 37.2 on the Noometry Index.

Last verified . 39 shared benchmarks.

GPT-4.5 OpenAI

37.2

Rank #208 Confirmed

o1 OpenAI

40.9

Rank #143 Confirmed

Summary

  • They share 39 benchmarks with published results for both. GPT-4.5 scores higher in 4 categories and o1 in 6 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where o1 leads 27.9 to 13.9.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for GPT-4.5 and 73.3% for o1.

Side by side

GPT-4.5o1
ProviderOpenAIOpenAI
Noometry Index37.240.9
Released2025-02-272024-09-12
WeightsProprietaryProprietary
Context window—200K
Max output—100K
Input $ / M tokens—$15
Output $ / M tokens—$60
Results tracked4252

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

Category by category

Coding o1 leads

GPT-4.5: 42.2 (#109), o1: 46.1 (#70)

BenchmarkGPT-4.5o1
Aider Polyglot44.9%61.7%
WeirdML39.4%47.6%
LiveBench Coding75.2%69.7%
LMArena Coding13961367
CadEval—56%
HumanEval+—89%
MBPP+—80.2%

Agentic & Tool Use GPT-4.5 leads

GPT-4.5: 27.9 (#97), o1: 24.6 (#117)

BenchmarkGPT-4.5o1
Cybench17.5%10%
METR Time Horizons—51.1%

Reasoning o1 leads

GPT-4.5: 13.9 (#330), o1: 27.9 (#111)

BenchmarkGPT-4.5o1
SimpleBench34.5%41.7%
ARC-AGI-110.3%30.7%
EnigmaEval3.2%5.7%
LiveBench Reasoning71.1%91.6%
LMArena Hard Prompts14031371
LiveBench Data Analysis64.3%65.5%
Epoch Capabilities Index136.74141.91
LiveBench69%75.7%
ARC-AGI-20.8%—
Chess Puzzles—15%
DTBench—74.7%
LMCA—22.3%
ForecastBench61.7—

Math o1 leads

GPT-4.5: 32.6 (#211), o1: 36.1 (#175)

BenchmarkGPT-4.5o1
OTIS Mock AIME 2024-202537.8%73.3%
LiveBench Math69.3%80.3%
LMArena Math14121388
MATH Level 578.6%94.7%
FrontierMath (Tiers 1-3)—14.7%
FrontierMath (Feb 2025 set)—9.3%

Knowledge o1 leads

GPT-4.5: 32.5 (#211), o1: 41.5 (#110)

BenchmarkGPT-4.5o1
GPQA Diamond68.7%76.8%
Humanity's Last Exam5.4%8%
Confabulations13.6%11.7%
LMArena Expert13941361
SimpleQA Verified—41.1%

Multimodal GPT-4.5 leads

GPT-4.5: 37.6 (#71), o1: 34.2 (#93)

BenchmarkGPT-4.5o1
LMArena Vision11951168
VPCT45%37%
GeoBench—80%
SpatialViz-Bench—41.4%

Multilingual GPT-4.5 leads

GPT-4.5: 52.5 (#83), o1: 48.6 (#142)

BenchmarkGPT-4.5o1
LMArena Non-English14131358
LMArena Chinese14211394
LMArena French14181344
LMArena German14571337
LMArena Japanese14161346
LMArena Korean13921396
LMArena Russian14191356
LMArena Spanish—1345

Instruction Following o1 leads

GPT-4.5: 72.6 (#134), o1: 74.8 (#86)

BenchmarkGPT-4.5o1
LiveBench Instruction Following72.3%81.5%
LMArena Instruction Following14041367

Long Context o1 leads

GPT-4.5: 40.4 (#155), o1: 50.3 (#9)

BenchmarkGPT-4.5o1
Fiction.LiveBench63.9%83.3%
LMArena Longer Query14061378

Writing & Preference GPT-4.5 leads

GPT-4.5: 56.9 (#134), o1: 55.6 (#144)

BenchmarkGPT-4.5o1
LMArena Text14171366
LMArena Creative Writing13941348
Short-Story Creative Writing75.6%70.2%
LMArena Multi-Turn14441369
LiveBench Language61.5%65.4%
EQ-Bench Creative Writing1258—

Frequently asked questions

Is GPT-4.5 better than o1?

o1 is the stronger model overall, scoring 40.9 to 37.2 on the Noometry Index.

Is GPT-4.5 or o1 better for coding?

o1 scores higher on coding benchmarks: 46.1 versus 42.2 in the Noometry coding category.

How many benchmarks do GPT-4.5 and o1 share?

39 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and o1 has 52.

Related comparisons

Go deeper