Noometry

Model comparison

GPT-4.5 vs GPT-5.2

GPT-5.2 is the stronger model overall, scoring 54.1 to 37.2 on the Noometry Index.

Last verified . 29 shared benchmarks.

GPT-4.5 OpenAI

37.2

Rank #208 Confirmed

GPT-5.2 OpenAI

54.1

Rank #34 Confirmed

Summary

  • They share 29 benchmarks with published results for both. GPT-4.5 scores higher in 0 categories and GPT-5.2 in 10 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where GPT-5.2 leads 50.2 to 13.9.
  • The biggest single-benchmark swing is ARC-AGI-1: 10.3% for GPT-4.5 and 86.2% for GPT-5.2.

Side by side

GPT-4.5GPT-5.2
ProviderOpenAIOpenAI
Noometry Index37.254.1
Released2025-02-272025-12-11
WeightsProprietaryProprietary
Context window—400K
Max output—128K
Input $ / M tokens—$1.75
Output $ / M tokens—$14
Results tracked4267

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

Category by category

Coding GPT-5.2 leads

GPT-4.5: 42.2 (#109), GPT-5.2: 51.6 (#37)

BenchmarkGPT-4.5GPT-5.2
WeirdML39.4%72.2%
LMArena Coding13961447
SWE-bench Verified—73.8%
SWE-bench Verified (bash only)—72.8%
Aider Polyglot44.9%—
LMArena WebDev—1416
SWE-bench Multilingual—66.7%
GSO—27.4%
LiveBench Coding75.2%—
ALE-Bench—1,294
AlgoTune—2.05

Agentic & Tool Use GPT-5.2 leads

GPT-4.5: 27.9 (#97), GPT-5.2: 40.2 (#24)

BenchmarkGPT-4.5GPT-5.2
Terminal-Bench—64.9%
Berkeley Function Calling Leaderboard—55.9%
GDPval—49.7%
Remote Labor Index—2.5%
τ²-bench Airline—83%
τ²-bench Banking—32.2%
τ²-bench Retail—81.6%
τ²-bench Telecom—89.7%
Cybench17.5%—
DeepResearch Bench—41.1%
LMArena Search—1207
METR Time Horizons—75.3%
Vending-Bench 2—3,591

Reasoning GPT-5.2 leads

GPT-4.5: 13.9 (#330), GPT-5.2: 50.2 (#35)

BenchmarkGPT-4.5GPT-5.2
ARC-AGI-20.8%52.9%
SimpleBench34.5%45.8%
ARC-AGI-110.3%86.2%
EnigmaEval3.2%10.4%
LMArena Hard Prompts14031445
Epoch Capabilities Index136.74153.45
ForecastBench61.760.1
Kagi LLM Benchmark—73.3%
NYT Connections (extended)—83.6%
Chess Puzzles—49%
EBR-Bench—23%
LiveBench Reasoning71.1%—
Mystery Game Puzzles—23%
DTBench—90.9%
LiveBench Data Analysis64.3%—
LMCA—43.9%
LiveBench69%—

Math GPT-5.2 leads

GPT-4.5: 32.6 (#211), GPT-5.2: 60.0 (#38)

BenchmarkGPT-4.5GPT-5.2
OTIS Mock AIME 2024-202537.8%96.1%
LMArena Math14121440
FrontierMath (Tiers 1-3)—67.4%
FrontierMath Tier 4—31.7%
MathArena Final-Answer Competitions—72%
ProofBench—15%
LiveBench Math69.3%—
MATH Level 578.6%—
FrontierMath (Feb 2025 set)—40.7%
FrontierMath Tier 4 (v1)—18.8%

Knowledge GPT-5.2 leads

GPT-4.5: 32.5 (#211), GPT-5.2: 59.3 (#32)

BenchmarkGPT-4.5GPT-5.2
GPQA Diamond68.7%91.4%
Humanity's Last Exam5.4%27.8%
LMArena Expert13941445
SimpleQA Verified—37.1%
Confabulations13.6%—
Vectara Hallucination Rate—8.4%

Multimodal GPT-5.2 leads

GPT-4.5: 37.6 (#71), GPT-5.2: 51.3 (#7)

BenchmarkGPT-4.5GPT-5.2
LMArena Vision11951268
VPCT45%84%
Furniture Assembly—38.3%
LMArena Document—1405

Multilingual Too close to call

GPT-4.5: 52.5 (#83), GPT-5.2: 53.4 (#67)

BenchmarkGPT-4.5GPT-5.2
LMArena Non-English14131425
LMArena Chinese14211460
LMArena French14181455
LMArena German14571448
LMArena Japanese14161420
LMArena Korean13921392
LMArena Russian14191440
LMArena Spanish—1433

Instruction Following GPT-5.2 leads

GPT-4.5: 72.6 (#134), GPT-5.2: 74.7 (#89)

BenchmarkGPT-4.5GPT-5.2
LMArena Instruction Following14041417
LiveBench Instruction Following72.3%—

Long Context GPT-5.2 leads

GPT-4.5: 40.4 (#155), GPT-5.2: 44.0 (#78)

BenchmarkGPT-4.5GPT-5.2
LMArena Longer Query14061428
Fiction.LiveBench63.9%—
CL-bench—18.2%

Writing & Preference GPT-5.2 leads

GPT-4.5: 56.9 (#134), GPT-5.2: 66.8 (#32)

BenchmarkGPT-4.5GPT-5.2
LMArena Text14171439
LMArena Creative Writing13941401
EQ-Bench Creative Writing12581703
LMArena Multi-Turn14441458
Short-Story Creative Writing75.6%—
LiveBench Language61.5%—

Frequently asked questions

Is GPT-4.5 better than GPT-5.2?

GPT-5.2 is the stronger model overall, scoring 54.1 to 37.2 on the Noometry Index.

Is GPT-4.5 or GPT-5.2 better for coding?

GPT-5.2 scores higher on coding benchmarks: 51.6 versus 42.2 in the Noometry coding category.

How many benchmarks do GPT-4.5 and GPT-5.2 share?

29 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and GPT-5.2 has 67.

Related comparisons

Go deeper