Noometry

Model comparison

GPT-5 Mini vs Grok 4

Grok 4 is the stronger model overall, scoring 48.1 to 41.8 on the Noometry Index.

Last verified . 39 shared benchmarks.

GPT-5 Mini OpenAI

41.8

Rank #128 Confirmed

Grok 4 xAI

48.1

Rank #56 Confirmed

Summary

  • They share 39 benchmarks with published results for both. GPT-5 Mini scores higher in 1 category and Grok 4 in 9 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in long context, where Grok 4 leads 63.1 to 41.9.
  • The biggest single-benchmark swing is Fiction.LiveBench: 69.4% for GPT-5 Mini and 94.4% for Grok 4.

Side by side

GPT-5 MiniGrok 4
ProviderOpenAIxAI
Noometry Index41.848.1
Released2025-08-072025-07-09
WeightsProprietaryProprietary
Context window400K—
Max output128K—
Input $ / M tokens$0.25—
Output $ / M tokens$2—
Results tracked6048

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

Category by category

Coding Grok 4 leads

GPT-5 Mini: 40.1 (#146), Grok 4: 50.3 (#46)

BenchmarkGPT-5 MiniGrok 4
WeirdML52.7%45.7%
LMArena Coding14061408
SWE-bench Verified64.7%—
SWE-bench Verified (bash only)59.8%—
Aider Polyglot—79.6%
SWE-bench Multilingual39.7%—
SciCode39.2%—
ALE-Bench799.77—
AlgoTune1.38—

Agentic & Tool Use Grok 4 leads

GPT-5 Mini: 31.1 (#70), Grok 4: 32.3 (#68)

BenchmarkGPT-5 MiniGrok 4
Terminal-Bench34.8%27.2%
Berkeley Function Calling Leaderboard55.5%63%
GDPval—21.1%
Cybench—43%
DeepResearch Bench—47.3%
BALROG—43.6%
LMArena Search—1142
METR Time Horizons—66.6%
Vending-Bench 2-31.18—

Reasoning Grok 4 leads

GPT-5 Mini: 23.9 (#168), Grok 4: 36.7 (#65)

BenchmarkGPT-5 MiniGrok 4
ARC-AGI-24.4%16%
Kagi LLM Benchmark70.3%73.6%
ARC-AGI-154.3%66.7%
Chess Puzzles30%28%
LMArena Hard Prompts13801409
Epoch Capabilities Index145.52146.44
ForecastBench6160.9
SimpleBench—60.5%
CritPt0%—
EnigmaEval8.2%—
Mystery Game Puzzles10%—
DTBench80.5%—
LMCA34.2%—

Math Grok 4 leads

GPT-5 Mini: 46.7 (#69), Grok 4: 48.4 (#64)

BenchmarkGPT-5 MiniGrok 4
OTIS Mock AIME 2024-202586.7%84%
Omni-MATH72.2%60.3%
LMArena Math13781422
FrontierMath (Feb 2025 set)27.2%19.7%
FrontierMath Tier 4 (v1)6.3%2.1%
FrontierMath (Tiers 1-3)46.7%—
FrontierMath Tier 412.2%—
ProofBench9%—
MATH Level 597.8%—

Knowledge Grok 4 leads

GPT-5 Mini: 45.6 (#86), Grok 4: 53.8 (#55)

BenchmarkGPT-5 MiniGrok 4
GPQA Diamond75%87%
MMLU-Pro83.5%85.1%
Confabulations13.3%12.4%
GPQA (HELM)75.6%72.7%
LMArena Expert13791415
Humanity's Last Exam19.4%—
SimpleQA Verified21.6%—
Vectara Hallucination Rate12.9%—

Multimodal GPT-5 Mini leads

GPT-5 Mini: 35.6 (#85), Grok 4: 33.7 (#94)

BenchmarkGPT-5 MiniGrok 4
LMArena Vision12021210
GeoBench—45%
VPCT40.2%—

Multilingual Grok 4 leads

GPT-5 Mini: 48.9 (#137), Grok 4: 51.8 (#103)

BenchmarkGPT-5 MiniGrok 4
LMArena Non-English13631403
LMArena Chinese13851427
LMArena French13861418
LMArena German13661429
LMArena Japanese13411394
LMArena Korean13081377
LMArena Russian13621410
LMArena Spanish13551420

Instruction Following Grok 4 leads

GPT-5 Mini: 76.2 (#46), Grok 4: 79.2 (#5)

BenchmarkGPT-5 MiniGrok 4
IFEval92.7%94.9%
LMArena Instruction Following13571387

Long Context Grok 4 leads

GPT-5 Mini: 41.9 (#132), Grok 4: 63.1 (#4)

BenchmarkGPT-5 MiniGrok 4
Fiction.LiveBench69.4%94.4%
LMArena Longer Query13551409

Writing & Preference Grok 4 leads

GPT-5 Mini: 55.2 (#148), Grok 4: 58.5 (#116)

BenchmarkGPT-5 MiniGrok 4
LMArena Text13731411
LMArena Creative Writing13251397
Short-Story Creative Writing83.1%76.9%
WildBench85.5%79.7%
LMArena Multi-Turn13631416
EQ-Bench Creative Writing1313—

Frequently asked questions

Is GPT-5 Mini better than Grok 4?

Grok 4 is the stronger model overall, scoring 48.1 to 41.8 on the Noometry Index.

Is GPT-5 Mini or Grok 4 better for coding?

Grok 4 scores higher on coding benchmarks: 50.3 versus 40.1 in the Noometry coding category.

How many benchmarks do GPT-5 Mini and Grok 4 share?

39 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Grok 4 has 48.

Related comparisons

Go deeper