Noometry

Model comparison

GPT-4 vs o3

o3 is the stronger model overall, scoring 47.5 to 29.1 on the Noometry Index.

Last verified . 29 shared benchmarks.

GPT-4 OpenAI

29.1

Rank #316 Confirmed

o3 OpenAI

47.5

Rank #61 Confirmed

Summary

  • They share 29 benchmarks with published results for both. GPT-4 scores higher in 0 categories and o3 in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where o3 leads 50.2 to 10.8.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.1% for GPT-4 and 84.4% for o3.
  • o3 is cheaper at $2 / $8 per million input/output tokens, against $30 / $60 for GPT-4.
  • o3 accepts more context: 200K tokens versus 8K.

Side by side

GPT-4o3
ProviderOpenAIOpenAI
Noometry Index29.147.5
Released2023-03-142025-04-16
WeightsProprietaryProprietary
Context window8K200K
Max output8K100K
Input $ / M tokens$30$2
Output $ / M tokens$60$8
Results tracked3863

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

Category by category

Coding o3 leads

GPT-4: 31.6 (#283), o3: 46.8 (#64)

BenchmarkGPT-4o3
WeirdML12.4%52.4%
LMArena Coding12541408
SWE-bench Verified—62.3%
SWE-bench Verified (bash only)—58.4%
Aider Polyglot—81.3%
GSO—8.8%
BigCodeBench Instruct46%—
BigCodeBench Complete57.2%—
CadEval—74%
ALE-Bench—933.55
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, o3: 34.5 (#44)

BenchmarkGPT-4o3
METR Time Horizons36.1%65.4%
Berkeley Function Calling Leaderboard—63%
GDPval—30.8%
DeepResearch Bench—45.2%
OSWorld—23%
LMArena Search—1144

Reasoning o3 leads

GPT-4: 17.8 (#289), o3: 32.0 (#78)

BenchmarkGPT-4o3
Chess Puzzles4%38%
LMArena Hard Prompts12411402
Mystery Game Puzzles12%29%
DTBench62.7%84.8%
LMCA17.1%39.7%
Epoch Capabilities Index125.89146.86
ForecastBench57.862.5
ARC-AGI-2—6.5%
SimpleBench—53.1%
Kagi LLM Benchmark—67.6%
ARC-AGI-1—60.8%
CritPt—1.4%
EnigmaEval—13.1%
BIG-Bench Hard75.1%—
HellaSwag95.3%—
WinoGrande87.5%—

Math o3 leads

GPT-4: 10.8 (#309), o3: 50.2 (#58)

BenchmarkGPT-4o3
OTIS Mock AIME 2024-20251.1%84.4%
LMArena Math12691426
MATH Level 523%97.8%
FrontierMath (Tiers 1-3)—33.3%
Omni-MATH—71.4%
FrontierMath (Feb 2025 set)—18.7%
FrontierMath Tier 4 (v1)—2.1%
GSM8K92%—

Knowledge o3 leads

GPT-4: 18.4 (#282), o3: 54.6 (#52)

BenchmarkGPT-4o3
GPQA Diamond35.7%81.8%
LMArena Expert12111402
Humanity's Last Exam—20.3%
SimpleQA Verified—49.4%
MMLU-Pro—85.9%
Confabulations—14.4%
GPQA (HELM)—75.3%
MMLU86.4%—
TriviaQA84.8%—

Multimodal Not comparable

GPT-4: —, o3: 41.4 (#36)

BenchmarkGPT-4o3
LMArena Vision—1214
GeoBench—74%
VPCT—52%

Multilingual o3 leads

GPT-4: 40.6 (#215), o3: 51.7 (#105)

BenchmarkGPT-4o3
LMArena Non-English12461401
LMArena Chinese12421437
LMArena French12831430
LMArena German12511420
LMArena Japanese12091403
LMArena Korean11841370
LMArena Russian12511406
LMArena Spanish12611395

Instruction Following o3 leads

GPT-4: 65.3 (#222), o3: 72.8 (#127)

BenchmarkGPT-4o3
LMArena Instruction Following12411368
IFEval—86.9%

Long Context o3 leads

GPT-4: 37.7 (#212), o3: 53.3 (#6)

BenchmarkGPT-4o3
LMArena Longer Query12441372
Fiction.LiveBench—88.9%
CL-bench—17.8%

Writing & Preference o3 leads

GPT-4: 34.9 (#268), o3: 63.5 (#64)

BenchmarkGPT-4o3
LMArena Text12631410
LMArena Creative Writing12441359
EQ-Bench Creative Writing7521676
LMArena Multi-Turn12571405
Short-Story Creative Writing—83.9%
WildBench—86.1%

Frequently asked questions

Is GPT-4 better than o3?

o3 is the stronger model overall, scoring 47.5 to 29.1 on the Noometry Index.

Which is cheaper, GPT-4 or o3?

o3 is cheaper. It lists at $2 per million input tokens and $8 per million output tokens; GPT-4 lists at $30 and $60.

Is GPT-4 or o3 better for coding?

o3 scores higher on coding benchmarks: 46.8 versus 31.6 in the Noometry coding category.

Which has the bigger context window?

o3 does, with 200K tokens against 8K.

How many benchmarks do GPT-4 and o3 share?

29 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and o3 has 63.

Related comparisons

Go deeper