Noometry

Model comparison

GPT-4 vs o3-mini

o3-mini is the stronger model overall, scoring 36.7 to 28.9 on the Noometry Index.

Last verified . 27 shared benchmarks.

GPT-4 OpenAI

28.9

Rank #308 Confirmed

o3-mini OpenAI

36.7

Rank #205 Confirmed

Summary

  • They share 27 benchmarks with published results for both. GPT-4 scores higher in 2 categories and o3-mini in 6 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where o3-mini leads 38.3 to 18.4.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.1% for GPT-4 and 76.9% for o3-mini.
  • o3-mini is cheaper at $1.10 / $4.40 per million input/output tokens, against $30 / $60 for GPT-4.
  • o3-mini accepts more context: 200K tokens versus 8K.

Side by side

GPT-4o3-mini
ProviderOpenAIOpenAI
Noometry Index28.936.7
Released2023-03-142024-12-20
WeightsProprietaryProprietary
Context window8K200K
Max output8K100K
Input $ / M tokens$30$1.10
Output $ / M tokens$60$4.40
Results tracked3851

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

Category by category

Coding o3-mini leads

GPT-4: 31.2 (#277), o3-mini: 41.4 (#116)

BenchmarkGPT-4o3-mini
WeirdML12.4%43.7%
LMArena Coding12541378
Aider Polyglot—60.4%
SciCode—39.8%
GSO—1.3%
BigCodeBench Instruct46%—
LiveBench Coding—82.7%
BigCodeBench Complete57.2%—
CadEval—54%
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, o3-mini: 29.3 (#79)

BenchmarkGPT-4o3-mini
Cybench—22.5%
METR Time Horizons36.1%—

Reasoning Too close to call

GPT-4: 17.3 (#285), o3-mini: 16.3 (#295)

BenchmarkGPT-4o3-mini
Chess Puzzles4%17%
LMArena Hard Prompts12411366
Mystery Game Puzzles12%7%
DTBench62.7%68.8%
LMCA17.1%19%
Epoch Capabilities Index125.88140.35
ForecastBench57.859.6
ARC-AGI-2—3%
SimpleBench—22.8%
ARC-AGI-1—34.5%
CritPt—0.3%
LiveBench Reasoning—89.6%
LiveBench Data Analysis—70.6%
BIG-Bench Hard75.1%—
HellaSwag95.3%—
LiveBench—75.9%
WinoGrande87.5%—

Math o3-mini leads

GPT-4: 10.1 (#304), o3-mini: 27.6 (#236)

BenchmarkGPT-4o3-mini
OTIS Mock AIME 2024-20251.1%76.9%
LMArena Math12691396
MATH Level 523%96.5%
FrontierMath (Tiers 1-3)—18.6%
FrontierMath Tier 4—0%
LiveBench Math—77.3%
FrontierMath (Feb 2025 set)—12.4%
FrontierMath Tier 4 (v1)—4.2%
GSM8K92%—

Knowledge o3-mini leads

GPT-4: 18.4 (#277), o3-mini: 38.3 (#135)

BenchmarkGPT-4o3-mini
GPQA Diamond35.7%77%
LMArena Expert12111364
SimpleQA Verified—15.3%
Confabulations—17.9%
MMLU86.4%—
TriviaQA84.8%—

Multilingual o3-mini leads

GPT-4: 40.8 (#210), o3-mini: 45.9 (#159)

BenchmarkGPT-4o3-mini
LMArena Non-English12461319
LMArena Chinese12421379
LMArena French12821333
LMArena German12511303
LMArena Japanese12091286
LMArena Korean11841314
LMArena Russian12511304
LMArena Spanish12611321

Instruction Following o3-mini leads

GPT-4: 65.5 (#217), o3-mini: 75.3 (#67)

BenchmarkGPT-4o3-mini
LMArena Instruction Following12411338
LiveBench Instruction Following—84.4%

Long Context GPT-4 leads

GPT-4: 37.7 (#207), o3-mini: 33.8 (#251)

BenchmarkGPT-4o3-mini
LMArena Longer Query12441343
Fiction.LiveBench—50%

Writing & Preference o3-mini leads

GPT-4: 35.0 (#263), o3-mini: 50.4 (#177)

BenchmarkGPT-4o3-mini
LMArena Text12631337
LMArena Creative Writing12441286
LMArena Multi-Turn12571320
Short-Story Creative Writing—61.7%
EQ-Bench Creative Writing752—
LiveBench Language—50.7%

Frequently asked questions

Is GPT-4 better than o3-mini?

o3-mini is the stronger model overall, scoring 36.7 to 28.9 on the Noometry Index.

Which is cheaper, GPT-4 or o3-mini?

o3-mini is cheaper. It lists at $1.10 per million input tokens and $4.40 per million output tokens; GPT-4 lists at $30 and $60.

Is GPT-4 or o3-mini better for coding?

o3-mini scores higher on coding benchmarks: 41.4 versus 31.2 in the Noometry coding category.

Which has the bigger context window?

o3-mini does, with 200K tokens against 8K.

How many benchmarks do GPT-4 and o3-mini share?

27 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and o3-mini has 51.

Related comparisons

Go deeper