Noometry

Model comparison

GPT-4.5 vs o4-mini

o4-mini is the stronger model overall, scoring 41.6 to 37.2 on the Noometry Index.

Last verified . 33 shared benchmarks.

GPT-4.5 OpenAI

37.2

Rank #208 Confirmed

o4-mini OpenAI

41.6

Rank #132 Confirmed

Summary

  • They share 33 benchmarks with published results for both. GPT-4.5 scores higher in 3 categories and o4-mini in 7 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where o4-mini leads 43.6 to 32.5.
  • The biggest single-benchmark swing is ARC-AGI-1: 10.3% for GPT-4.5 and 58.7% for o4-mini.

Side by side

GPT-4.5o4-mini
ProviderOpenAIOpenAI
Noometry Index37.241.6
Released2025-02-272025-04-16
WeightsProprietaryProprietary
Context window—200K
Max output—100K
Input $ / M tokens—$1.10
Output $ / M tokens—$4.40
Results tracked4260

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

Category by category

Coding GPT-4.5 leads

GPT-4.5: 42.2 (#109), o4-mini: 40.9 (#127)

BenchmarkGPT-4.5o4-mini
Aider Polyglot44.9%72%
WeirdML39.4%52.6%
LMArena Coding13961368
SWE-bench Verified (bash only)—45%
GSO—3.6%
LiveBench Coding75.2%—
CadEval—62%
ALE-Bench—826.17
AlgoTune—1.72

Agentic & Tool Use o4-mini leads

GPT-4.5: 27.9 (#97), o4-mini: 32.6 (#61)

BenchmarkGPT-4.5o4-mini
Berkeley Function Calling Leaderboard—53.2%
GDPval—25.3%
Cybench17.5%—
METR Time Horizons—63.9%

Reasoning o4-mini leads

GPT-4.5: 13.9 (#330), o4-mini: 24.6 (#162)

BenchmarkGPT-4.5o4-mini
ARC-AGI-20.8%6.1%
SimpleBench34.5%38.7%
ARC-AGI-110.3%58.7%
EnigmaEval3.2%9.2%
LMArena Hard Prompts14031351
Epoch Capabilities Index136.74145.64
ForecastBench61.761.8
Kagi LLM Benchmark—67.6%
CritPt—0.6%
Chess Puzzles—26%
LiveBench Reasoning71.1%—
Mystery Game Puzzles—5%
DTBench—77.6%
LiveBench Data Analysis64.3%—
LMCA—26.5%
LiveBench69%—

Math o4-mini leads

GPT-4.5: 32.6 (#211), o4-mini: 40.8 (#89)

BenchmarkGPT-4.5o4-mini
OTIS Mock AIME 2024-202537.8%81.7%
LMArena Math14121389
MATH Level 578.6%97.8%
FrontierMath (Tiers 1-3)—36.1%
FrontierMath Tier 4—4.9%
Omni-MATH—72%
LiveBench Math69.3%—
FrontierMath (Feb 2025 set)—24.8%
FrontierMath Tier 4 (v1)—6.3%

Knowledge o4-mini leads

GPT-4.5: 32.5 (#211), o4-mini: 43.6 (#91)

BenchmarkGPT-4.5o4-mini
GPQA Diamond68.7%79.6%
Humanity's Last Exam5.4%18.1%
Confabulations13.6%15.8%
LMArena Expert13941343
SimpleQA Verified—19.6%
MMLU-Pro—82%
Vectara Hallucination Rate—18.6%
GPQA (HELM)—73.5%

Multimodal o4-mini leads

GPT-4.5: 37.6 (#71), o4-mini: 40.2 (#49)

BenchmarkGPT-4.5o4-mini
LMArena Vision11951194
VPCT45%57.5%
GeoBench—64%

Multilingual GPT-4.5 leads

GPT-4.5: 52.5 (#83), o4-mini: 47.0 (#154)

BenchmarkGPT-4.5o4-mini
LMArena Non-English14131337
LMArena Chinese14211354
LMArena French14181364
LMArena German14571336
LMArena Japanese14161308
LMArena Korean13921312
LMArena Russian14191334
LMArena Spanish—1347

Instruction Following o4-mini leads

GPT-4.5: 72.6 (#134), o4-mini: 75.2 (#68)

BenchmarkGPT-4.5o4-mini
LMArena Instruction Following14041321
LiveBench Instruction Following72.3%—
IFEval—92.8%

Long Context o4-mini leads

GPT-4.5: 40.4 (#155), o4-mini: 45.5 (#33)

BenchmarkGPT-4.5o4-mini
Fiction.LiveBench63.9%77.8%
LMArena Longer Query14061315

Writing & Preference GPT-4.5 leads

GPT-4.5: 56.9 (#134), o4-mini: 54.0 (#152)

BenchmarkGPT-4.5o4-mini
LMArena Text14171353
LMArena Creative Writing13941294
Short-Story Creative Writing75.6%75%
LMArena Multi-Turn14441350
EQ-Bench Creative Writing1258—
WildBench—85.4%
LiveBench Language61.5%—

Frequently asked questions

Is GPT-4.5 better than o4-mini?

o4-mini is the stronger model overall, scoring 41.6 to 37.2 on the Noometry Index.

Is GPT-4.5 or o4-mini better for coding?

GPT-4.5 scores higher on coding benchmarks: 42.2 versus 40.9 in the Noometry coding category.

How many benchmarks do GPT-4.5 and o4-mini share?

33 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and o4-mini has 60.

Related comparisons

Go deeper