Noometry

Model comparison

GPT-5.4 mini vs Grok-3 mini

GPT-5.4 mini is the stronger model overall, scoring 45.0 to 41.2 on the Noometry Index.

Last verified . 25 shared benchmarks.

GPT-5.4 mini OpenAI

45.0

Rank #76 Confirmed

Grok-3 mini xAI

41.2

Rank #141 Confirmed

Summary

  • They share 25 benchmarks with published results for both. GPT-5.4 mini scores higher in 7 categories and Grok-3 mini in 1 category; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where GPT-5.4 mini leads 30.4 to 13.6.
  • The biggest single-benchmark swing is ARC-AGI-1: 63.7% for GPT-5.4 mini and 16.5% for Grok-3 mini.

Side by side

GPT-5.4 miniGrok-3 mini
ProviderOpenAIxAI
Noometry Index45.041.2
Released2026-03-172025-04-09
WeightsProprietaryProprietary
Context window400K—
Max output128K—
Input $ / M tokens$0.75—
Output $ / M tokens$4.50—
Results tracked4635

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

Category by category

Coding GPT-5.4 mini leads

GPT-5.4 mini: 45.2 (#72), Grok-3 mini: 40.8 (#131)

BenchmarkGPT-5.4 miniGrok-3 mini
WeirdML60.3%42.6%
LMArena Coding14381379
FrontierCode27%—
Aider Polyglot—49.3%
LMArena WebDev1397—
SciCode49.9%—
ALE-Bench1,189—

Agentic & Tool Use Not comparable

GPT-5.4 mini: 29.9 (#81), Grok-3 mini: —

BenchmarkGPT-5.4 miniGrok-3 mini
DeepResearch Bench36.3%—

Reasoning GPT-5.4 mini leads

GPT-5.4 mini: 30.4 (#85), Grok-3 mini: 13.6 (#334)

BenchmarkGPT-5.4 miniGrok-3 mini
ARC-AGI-218.9%0.4%
Kagi LLM Benchmark37.9%61.3%
ARC-AGI-163.7%16.5%
LMArena Hard Prompts14241375
Epoch Capabilities Index148.84140.35
NYT Connections (extended)61.8%—
CritPt10%—
Chess Puzzles24%—
Thematic Generalization61.7%—
Mystery Game Puzzles11%—
DTBench80%—
LMCA40.8%—
ForecastBench57—

Math GPT-5.4 mini leads

GPT-5.4 mini: 45.5 (#75), Grok-3 mini: 42.1 (#85)

BenchmarkGPT-5.4 miniGrok-3 mini
OTIS Mock AIME 2024-202588.9%77.8%
LMArena Math14191386
FrontierMath (Feb 2025 set)28.3%5.9%
FrontierMath (Tiers 1-3)51.2%—
FrontierMath Tier 49.8%—
ProofBench21%—
Omni-MATH—31.8%
MATH Level 5—90.9%
FrontierMath Tier 4 (v1)2.1%—

Knowledge GPT-5.4 mini leads

GPT-5.4 mini: 51.5 (#67), Grok-3 mini: 46.4 (#81)

BenchmarkGPT-5.4 miniGrok-3 mini
GPQA Diamond86.9%76.3%
LMArena Expert14351395
SimpleQA Verified29.4%—
MMLU-Pro—79.9%
Confabulations—10.8%
Vectara Hallucination Rate5.5%—
GPQA (HELM)—67.5%

Multimodal Not comparable

GPT-5.4 mini: 39.7 (#56), Grok-3 mini: —

BenchmarkGPT-5.4 miniGrok-3 mini
LMArena Vision1245—

Multilingual GPT-5.4 mini leads

GPT-5.4 mini: 51.9 (#96), Grok-3 mini: 48.1 (#145)

BenchmarkGPT-5.4 miniGrok-3 mini
LMArena Non-English14051352
LMArena Chinese14461387
LMArena French14401357
LMArena German14091349
LMArena Japanese13741342
LMArena Korean13681335
LMArena Russian14171353
LMArena Spanish14051381

Instruction Following Grok-3 mini leads

GPT-5.4 mini: 74.1 (#102), Grok-3 mini: 78.5 (#9)

BenchmarkGPT-5.4 miniGrok-3 mini
LMArena Instruction Following14051357
IFEval—95.1%

Long Context GPT-5.4 mini leads

GPT-5.4 mini: 43.0 (#112), Grok-3 mini: 41.0 (#147)

BenchmarkGPT-5.4 miniGrok-3 mini
LMArena Longer Query14071372
Fiction.LiveBench—66.7%

Writing & Preference GPT-5.4 mini leads

GPT-5.4 mini: 64.0 (#58), Grok-3 mini: 52.5 (#169)

BenchmarkGPT-5.4 miniGrok-3 mini
LMArena Text14121370
LMArena Creative Writing13701342
LMArena Multi-Turn14291355
Short-Story Creative Writing—73.5%
EQ-Bench Creative Writing1665—
WildBench—65.1%

Frequently asked questions

Is GPT-5.4 mini better than Grok-3 mini?

GPT-5.4 mini is the stronger model overall, scoring 45.0 to 41.2 on the Noometry Index.

Is GPT-5.4 mini or Grok-3 mini better for coding?

GPT-5.4 mini scores higher on coding benchmarks: 45.2 versus 40.8 in the Noometry coding category.

How many benchmarks do GPT-5.4 mini and Grok-3 mini share?

25 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Grok-3 mini has 35.

Related comparisons

Go deeper