Noometry

Model comparison

GPT-5.4 mini vs Grok 4 Fast

GPT-5.4 mini is the stronger model overall, scoring 45.0 to 39.4 on the Noometry Index.

Last verified . 26 shared benchmarks.

GPT-5.4 mini OpenAI

45.0

Rank #76 Confirmed

Grok 4 Fast xAI

39.4

Rank #167 Confirmed

Summary

  • They share 26 benchmarks with published results for both. GPT-5.4 mini scores higher in 8 categories and Grok 4 Fast in 1 category; 6 gaps are clear of the uncertainty.
  • The widest gap is in long context, where Grok 4 Fast leads 63.2 to 43.0.
  • The biggest single-benchmark swing is Kagi LLM Benchmark: 37.9% for GPT-5.4 mini and 66.1% for Grok 4 Fast.

Side by side

GPT-5.4 miniGrok 4 Fast
ProviderOpenAIxAI
Noometry Index45.039.4
Released2026-03-172025-09-19
WeightsProprietaryProprietary
Context window400K—
Max output128K—
Input $ / M tokens$0.75—
Output $ / M tokens$4.50—
Results tracked4630

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 4 Fast: 32.5 (#271)

BenchmarkGPT-5.4 miniGrok 4 Fast
LMArena WebDev13971159
WeirdML60.3%42.9%
LMArena Coding14381429
FrontierCode27%—
SciCode49.9%—
ALE-Bench1,189—

Agentic & Tool Use Too close to call

GPT-5.4 mini: 29.9 (#81), Grok 4 Fast: 29.5 (#86)

BenchmarkGPT-5.4 miniGrok 4 Fast
τ²-bench Banking—15.7%
Cybench—30%
DeepResearch Bench36.3%—
LMArena Search—1171

Reasoning GPT-5.4 mini leads

GPT-5.4 mini: 30.4 (#85), Grok 4 Fast: 22.2 (#201)

BenchmarkGPT-5.4 miniGrok 4 Fast
ARC-AGI-218.9%5.3%
Kagi LLM Benchmark37.9%66.1%
ARC-AGI-163.7%48.5%
LMArena Hard Prompts14241412
DTBench80%82.7%
Epoch Capabilities Index148.84144.2
ForecastBench5760.5
NYT Connections (extended)61.8%—
CritPt10%—
Chess Puzzles24%—
Thematic Generalization61.7%—
Mystery Game Puzzles11%—
LMCA40.8%—

Math GPT-5.4 mini leads

GPT-5.4 mini: 45.5 (#75), Grok 4 Fast: 38.9 (#123)

BenchmarkGPT-5.4 miniGrok 4 Fast
LMArena Math14191419
FrontierMath (Tiers 1-3)51.2%—
FrontierMath Tier 49.8%—
OTIS Mock AIME 2024-202588.9%—
ProofBench21%—
FrontierMath (Feb 2025 set)28.3%—
FrontierMath Tier 4 (v1)2.1%—

Knowledge GPT-5.4 mini leads

GPT-5.4 mini: 51.5 (#67), Grok 4 Fast: 32.0 (#214)

BenchmarkGPT-5.4 miniGrok 4 Fast
Vectara Hallucination Rate5.5%19.7%
LMArena Expert14351411
GPQA Diamond86.9%—
SimpleQA Verified29.4%—

Multimodal Not comparable

GPT-5.4 mini: 39.7 (#56), Grok 4 Fast: —

BenchmarkGPT-5.4 miniGrok 4 Fast
LMArena Vision1245—

Multilingual Too close to call

GPT-5.4 mini: 51.9 (#96), Grok 4 Fast: 51.3 (#111)

BenchmarkGPT-5.4 miniGrok 4 Fast
LMArena Non-English14051396
LMArena Chinese14461457
LMArena French14401431
LMArena German14091383
LMArena Japanese13741352
LMArena Korean13681357
LMArena Russian14171389
LMArena Spanish14051415

Instruction Following Too close to call

GPT-5.4 mini: 74.1 (#102), Grok 4 Fast: 73.2 (#121)

BenchmarkGPT-5.4 miniGrok 4 Fast
LMArena Instruction Following14051387

Long Context Grok 4 Fast leads

GPT-5.4 mini: 43.0 (#112), Grok 4 Fast: 63.2 (#3)

BenchmarkGPT-5.4 miniGrok 4 Fast
LMArena Longer Query14071415
Fiction.LiveBench—94.4%

Writing & Preference GPT-5.4 mini leads

GPT-5.4 mini: 64.0 (#58), Grok 4 Fast: 60.0 (#102)

BenchmarkGPT-5.4 miniGrok 4 Fast
LMArena Text14121407
LMArena Creative Writing13701387
LMArena Multi-Turn14291414
EQ-Bench Creative Writing1665—

Frequently asked questions

Is GPT-5.4 mini better than Grok 4 Fast?

GPT-5.4 mini is the stronger model overall, scoring 45.0 to 39.4 on the Noometry Index.

Is GPT-5.4 mini or Grok 4 Fast better for coding?

GPT-5.4 mini scores higher on coding benchmarks: 45.2 versus 32.5 in the Noometry coding category.

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

26 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Grok 4 Fast has 30.

Related comparisons

Go deeper