Noometry

Model comparison

GPT-4o mini vs GPT-5.4

GPT-5.4 is the stronger model overall, scoring 59.4 to 25.5 on the Noometry Index. GPT-4o mini costs 21× less per token, which makes it the better buy when GPT-5.4's lead doesn't matter for your workload.

Last verified . 31 shared benchmarks.

GPT-4o mini OpenAI

25.5

Rank #343 Confirmed

GPT-5.4 OpenAI

59.4

Rank #16 Confirmed

Summary

  • They share 31 benchmarks with published results for both. GPT-4o mini scores higher in 0 categories and GPT-5.4 in 10 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-5.4 leads 73.5 to 10.4.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.9% for GPT-4o mini and 97.8% for GPT-5.4.
  • GPT-4o mini is cheaper at $0.15 / $0.60 per million input/output tokens, against $2.50 / $15 for GPT-5.4.
  • GPT-5.4 accepts more context: 1.05M tokens versus 128K.

Side by side

GPT-4o miniGPT-5.4
ProviderOpenAIOpenAI
Noometry Index25.559.4
Released2024-07-182026-03-05
WeightsProprietaryProprietary
Context window128K1.05M
Max output16K128K
Input $ / M tokens$0.15$2.50
Output $ / M tokens$0.60$15
Results tracked6068

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

Category by category

Coding GPT-5.4 leads

GPT-4o mini: 22.0 (#335), GPT-5.4: 52.6 (#33)

BenchmarkGPT-4o miniGPT-5.4
WeirdML11.8%77.7%
LMArena Coding12901497
SWE-bench Verified—76.9%
DeepSWE—51.8%
Aider Polyglot3.6%—
LMArena WebDev—1465
SciCode—56.6%
GSO—31.4%
BigCodeBench Instruct46.1%—
LiveBench Coding43.1%—
MirrorCode—15.6%
BigCodeBench Complete57.4%—
ALE-Bench—1,607
AlgoTune—1.85
HumanEval+83.5%—
MBPP+72.2%—

Agentic & Tool Use GPT-5.4 leads

GPT-4o mini: 27.5 (#101), GPT-5.4: 46.5 (#13)

BenchmarkGPT-4o miniGPT-5.4
Terminal-Bench—81.8%
APEX-Agents—52.4%
τ²-bench Banking—39.4%
DeepResearch Bench—35.1%
PostTrainBench—19%
BALROG17.4%—
GBAEval—45.1%
LMArena Search—1197
METR Time Horizons—74.3%
Vending-Bench 2—6,144

Reasoning GPT-5.4 leads

GPT-4o mini: 8.7 (#347), GPT-5.4: 61.8 (#19)

BenchmarkGPT-4o miniGPT-5.4
ARC-AGI-20%74%
Kagi LLM Benchmark28.8%63.8%
Chess Puzzles0%44%
LMArena Hard Prompts12671485
Mystery Game Puzzles12%37%
DTBench54.4%94.4%
LMCA10.4%52%
Epoch Capabilities Index126.56156.81
SimpleBench10.7%—
NYT Connections (extended)—91.3%
ARC-AGI-1—93.7%
CritPt—23.4%
EnigmaEval—16%
Thematic Generalization—80%
EBR-Bench—25.4%
LiveBench Reasoning32.8%—
LiveBench Data Analysis50%—
ForecastBench—59.5
LiveBench41.3%—
PIQA88.7%—

Math GPT-5.4 leads

GPT-4o mini: 10.4 (#314), GPT-5.4: 73.5 (#19)

BenchmarkGPT-4o miniGPT-5.4
FrontierMath (Tiers 1-3)0.7%78.6%
OTIS Mock AIME 2024-20256.9%97.8%
LMArena Math12671488
FrontierMath Tier 4—49%
MathArena Final-Answer Competitions—83.1%
ProofBench—56%
Omni-MATH28%—
LiveBench Math36.3%—
MATH Level 552.6%—
FrontierMath (Feb 2025 set)—47.6%
FrontierMath Tier 4 (v1)—27.1%
GSM8K91.3%—

Knowledge GPT-5.4 leads

GPT-4o mini: 17.7 (#284), GPT-5.4: 65.3 (#14)

BenchmarkGPT-4o miniGPT-5.4
GPQA Diamond37.7%93.3%
SimpleQA Verified8.3%45.1%
LMArena Expert12351507
Humanity's Last Exam—36.2%
MMLU-Pro60.3%—
Confabulations37.2%—
Vectara Hallucination Rate—7%
GPQA (HELM)36.8%—
BoolQ88.7%—
MMLU81.8%—

Multimodal GPT-5.4 leads

GPT-4o mini: 25.9 (#122), GPT-5.4: 43.7 (#20)

BenchmarkGPT-4o miniGPT-5.4
LMArena Vision10661303
Video-MME64.8%—
GeoBench64%—
VPCT34%—
Blueprint-Bench 2—27.1%
Furniture Assembly—37.5%
LMArena Document—1471

Multilingual GPT-5.4 leads

GPT-4o mini: 42.0 (#199), GPT-5.4: 56.2 (#23)

BenchmarkGPT-4o miniGPT-5.4
LMArena Non-English12661465
LMArena Chinese12651519
LMArena French12971493
LMArena German12721472
LMArena Japanese12161485
LMArena Korean11951448
LMArena Russian12751480
LMArena Spanish12761454

Instruction Following GPT-5.4 leads

GPT-4o mini: 61.9 (#239), GPT-5.4: 77.1 (#27)

BenchmarkGPT-4o miniGPT-5.4
LMArena Instruction Following12581469
LiveBench Instruction Following56.8%—
IFEval78.2%—

Long Context GPT-5.4 leads

GPT-4o mini: 39.1 (#186), GPT-5.4: 50.3 (#8)

BenchmarkGPT-4o miniGPT-5.4
LMArena Longer Query12891473
CL-bench—27.9%
CL-bench Life—21.7%

Writing & Preference GPT-5.4 leads

GPT-4o mini: 39.5 (#248), GPT-5.4: 71.9 (#17)

BenchmarkGPT-4o miniGPT-5.4
LMArena Text12861469
LMArena Creative Writing12681439
EQ-Bench Creative Writing8731840
LMArena Multi-Turn12851482
Short-Story Creative Writing67.2%—
WildBench79.1%—
EQ-Bench 4—1272
LiveBench Language28.6%—

Frequently asked questions

Is GPT-4o mini better than GPT-5.4?

GPT-5.4 is the stronger model overall, scoring 59.4 to 25.5 on the Noometry Index. GPT-4o mini costs 21× less per token, which makes it the better buy when GPT-5.4's lead doesn't matter for your workload.

Which is cheaper, GPT-4o mini or GPT-5.4?

GPT-4o mini is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; GPT-5.4 lists at $2.50 and $15.

Is GPT-4o mini or GPT-5.4 better for coding?

GPT-5.4 scores higher on coding benchmarks: 52.6 versus 22.0 in the Noometry coding category.

Which has the bigger context window?

GPT-5.4 does, with 1.05M tokens against 128K.

How many benchmarks do GPT-4o mini and GPT-5.4 share?

31 benchmarks have published results for both models. GPT-4o mini has 60 scored results on Noometry and GPT-5.4 has 68.

Related comparisons

Go deeper