Noometry

Model comparison

GPT-4.5 vs GPT-5

GPT-5 is the stronger model overall, scoring 50.9 to 37.2 on the Noometry Index.

Last verified . 34 shared benchmarks.

GPT-4.5 OpenAI

37.2

Rank #208 Confirmed

GPT-5 OpenAI

50.9

Rank #45 Confirmed

Summary

  • They share 34 benchmarks with published results for both. GPT-4.5 scores higher in 1 category and GPT-5 in 9 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in long context, where GPT-5 leads 69.5 to 40.4.
  • The biggest single-benchmark swing is ARC-AGI-1: 10.3% for GPT-4.5 and 65.7% for GPT-5.

Side by side

GPT-4.5GPT-5
ProviderOpenAIOpenAI
Noometry Index37.250.9
Released2025-02-272025-08-07
WeightsProprietaryProprietary
Context window—400K
Max output—128K
Input $ / M tokens—$1.25
Output $ / M tokens—$10
Results tracked4269

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

Category by category

Coding GPT-5 leads

GPT-4.5: 42.2 (#109), GPT-5: 50.3 (#47)

BenchmarkGPT-4.5GPT-5
Aider Polyglot44.9%88%
WeirdML39.4%60.7%
LMArena Coding13961436
SWE-bench Verified—73.6%
SWE-bench Verified (bash only)—65%
LMArena WebDev—1418
SciCode—42.9%
GSO—6.9%
LiveBench Coding75.2%—
ALE-Bench—1,162
AlgoTune—1.67

Agentic & Tool Use GPT-5 leads

GPT-4.5: 27.9 (#97), GPT-5: 33.1 (#56)

BenchmarkGPT-4.5GPT-5
Terminal-Bench—49.6%
GDPval—34.8%
Remote Labor Index—1.7%
Cybench17.5%—
DeepResearch Bench—49.6%
BALROG—32.8%
LMArena Search—1133
METR Time Horizons—69.6%

Reasoning GPT-5 leads

GPT-4.5: 13.9 (#330), GPT-5: 38.3 (#64)

BenchmarkGPT-4.5GPT-5
ARC-AGI-20.8%9.9%
SimpleBench34.5%56.7%
ARC-AGI-110.3%65.7%
EnigmaEval3.2%10.5%
LMArena Hard Prompts14031416
Epoch Capabilities Index136.74150
ForecastBench61.761.4
Kagi LLM Benchmark—72.7%
CritPt—12.6%
Chess Puzzles—37%
EBR-Bench—12.7%
LiveBench Reasoning71.1%—
Mystery Game Puzzles—23%
DTBench—90.7%
LiveBench Data Analysis64.3%—
LMCA—40%
LiveBench69%—

Math GPT-5 leads

GPT-4.5: 32.6 (#211), GPT-5: 55.0 (#44)

BenchmarkGPT-4.5GPT-5
OTIS Mock AIME 2024-202537.8%91.4%
LMArena Math14121407
MATH Level 578.6%98.1%
FrontierMath (Tiers 1-3)—55.4%
FrontierMath Tier 4—22%
ProofBench—18%
Omni-MATH—64.7%
LiveBench Math69.3%—
FrontierMath (Feb 2025 set)—32.4%
FrontierMath Tier 4 (v1)—12.5%

Knowledge GPT-5 leads

GPT-4.5: 32.5 (#211), GPT-5: 56.6 (#43)

BenchmarkGPT-4.5GPT-5
GPQA Diamond68.7%86.2%
Humanity's Last Exam5.4%25.3%
Confabulations13.6%10.3%
LMArena Expert13941419
SimpleQA Verified—50.1%
MMLU-Pro—86.3%
Vectara Hallucination Rate—14.7%
GPQA (HELM)—79.2%

Multimodal GPT-5 leads

GPT-4.5: 37.6 (#71), GPT-5: 46.8 (#13)

BenchmarkGPT-4.5GPT-5
LMArena Vision11951232
VPCT45%66%
GeoBench—81%

Multilingual GPT-4.5 leads

GPT-4.5: 52.5 (#83), GPT-5: 51.4 (#110)

BenchmarkGPT-4.5GPT-5
LMArena Non-English14131397
LMArena Chinese14211422
LMArena French14181410
LMArena German14571416
LMArena Japanese14161409
LMArena Korean13921360
LMArena Russian14191406
LMArena Spanish—1399

Instruction Following GPT-5 leads

GPT-4.5: 72.6 (#134), GPT-5: 73.8 (#113)

BenchmarkGPT-4.5GPT-5
LMArena Instruction Following14041388
LiveBench Instruction Following72.3%—
IFEval—87.5%

Long Context GPT-5 leads

GPT-4.5: 40.4 (#155), GPT-5: 69.5 (#2)

BenchmarkGPT-4.5GPT-5
Fiction.LiveBench63.9%97.2%
LMArena Longer Query14061399

Writing & Preference GPT-5 leads

GPT-4.5: 56.9 (#134), GPT-5: 63.4 (#65)

BenchmarkGPT-4.5GPT-5
LMArena Text14171406
LMArena Creative Writing13941365
Short-Story Creative Writing75.6%86%
EQ-Bench Creative Writing12581627
LMArena Multi-Turn14441426
WildBench—85.7%
LiveBench Language61.5%—

Frequently asked questions

Is GPT-4.5 better than GPT-5?

GPT-5 is the stronger model overall, scoring 50.9 to 37.2 on the Noometry Index.

Is GPT-4.5 or GPT-5 better for coding?

GPT-5 scores higher on coding benchmarks: 50.3 versus 42.2 in the Noometry coding category.

How many benchmarks do GPT-4.5 and GPT-5 share?

34 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and GPT-5 has 69.

Related comparisons

Go deeper