Noometry

Model comparison

Grok-3 mini vs Grok 4

Grok 4 is the stronger model overall, scoring 48.1 to 41.2 on the Noometry Index.

Last verified . 34 shared benchmarks.

Grok-3 mini xAI

41.2

Rank #141 Confirmed

Grok 4 xAI

48.1

Rank #56 Confirmed

Summary

  • They share 34 benchmarks with published results for both. Grok-3 mini scores higher in 0 categories and Grok 4 in 8 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Grok 4 leads 36.7 to 13.6.
  • The biggest single-benchmark swing is ARC-AGI-1: 16.5% for Grok-3 mini and 66.7% for Grok 4.

Side by side

Grok-3 miniGrok 4
ProviderxAIxAI
Noometry Index41.248.1
Released2025-04-092025-07-09
WeightsProprietaryProprietary
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked3548

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

Category by category

Coding Grok 4 leads

Grok-3 mini: 40.8 (#131), Grok 4: 50.3 (#46)

BenchmarkGrok-3 miniGrok 4
Aider Polyglot49.3%79.6%
WeirdML42.6%45.7%
LMArena Coding13791408

Agentic & Tool Use Not comparable

Grok-3 mini: —, Grok 4: 32.3 (#68)

BenchmarkGrok-3 miniGrok 4
Terminal-Bench—27.2%
Berkeley Function Calling Leaderboard—63%
GDPval—21.1%
Cybench—43%
DeepResearch Bench—47.3%
BALROG—43.6%
LMArena Search—1142
METR Time Horizons—66.6%

Reasoning Grok 4 leads

Grok-3 mini: 13.6 (#334), Grok 4: 36.7 (#65)

BenchmarkGrok-3 miniGrok 4
ARC-AGI-20.4%16%
Kagi LLM Benchmark61.3%73.6%
ARC-AGI-116.5%66.7%
LMArena Hard Prompts13751409
Epoch Capabilities Index140.35146.44
SimpleBench—60.5%
Chess Puzzles—28%
ForecastBench—60.9

Math Grok 4 leads

Grok-3 mini: 42.1 (#85), Grok 4: 48.4 (#64)

BenchmarkGrok-3 miniGrok 4
OTIS Mock AIME 2024-202577.8%84%
Omni-MATH31.8%60.3%
LMArena Math13861422
FrontierMath (Feb 2025 set)5.9%19.7%
MATH Level 590.9%—
FrontierMath Tier 4 (v1)—2.1%

Knowledge Grok 4 leads

Grok-3 mini: 46.4 (#81), Grok 4: 53.8 (#55)

BenchmarkGrok-3 miniGrok 4
GPQA Diamond76.3%87%
MMLU-Pro79.9%85.1%
Confabulations10.8%12.4%
GPQA (HELM)67.5%72.7%
LMArena Expert13951415

Multimodal Not comparable

Grok-3 mini: —, Grok 4: 33.7 (#94)

BenchmarkGrok-3 miniGrok 4
LMArena Vision—1210
GeoBench—45%

Multilingual Grok 4 leads

Grok-3 mini: 48.1 (#145), Grok 4: 51.8 (#103)

BenchmarkGrok-3 miniGrok 4
LMArena Non-English13521403
LMArena Chinese13871427
LMArena French13571418
LMArena German13491429
LMArena Japanese13421394
LMArena Korean13351377
LMArena Russian13531410
LMArena Spanish13811420

Instruction Following Too close to call

Grok-3 mini: 78.5 (#9), Grok 4: 79.2 (#5)

BenchmarkGrok-3 miniGrok 4
IFEval95.1%94.9%
LMArena Instruction Following13571387

Long Context Grok 4 leads

Grok-3 mini: 41.0 (#147), Grok 4: 63.1 (#4)

BenchmarkGrok-3 miniGrok 4
Fiction.LiveBench66.7%94.4%
LMArena Longer Query13721409

Writing & Preference Grok 4 leads

Grok-3 mini: 52.5 (#169), Grok 4: 58.5 (#116)

BenchmarkGrok-3 miniGrok 4
LMArena Text13701411
LMArena Creative Writing13421397
Short-Story Creative Writing73.5%76.9%
WildBench65.1%79.7%
LMArena Multi-Turn13551416

Frequently asked questions

Is Grok-3 mini better than Grok 4?

Grok 4 is the stronger model overall, scoring 48.1 to 41.2 on the Noometry Index.

Is Grok-3 mini or Grok 4 better for coding?

Grok 4 scores higher on coding benchmarks: 50.3 versus 40.8 in the Noometry coding category.

How many benchmarks do Grok-3 mini and Grok 4 share?

34 benchmarks have published results for both models. Grok-3 mini has 35 scored results on Noometry and Grok 4 has 48.

Related comparisons

Go deeper