Noometry

Model comparison

Grok-3 mini vs Grok 4.6

Grok 4.6 is the stronger model overall, scoring 56.8 to 41.0 on the Noometry Index.

Last verified . 23 shared benchmarks.

Grok-3 mini xAI

41.0

Rank #130 Confirmed

Grok 4.6 xAI

56.8

Rank #19 Confirmed

Summary

  • They share 23 benchmarks with published results for both. Grok-3 mini scores higher in 1 category and Grok 4.6 in 7 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Grok 4.6 leads 61.3 to 13.8.
  • The biggest single-benchmark swing is ARC-AGI-1: 16.5% for Grok-3 mini and 87.5% for Grok 4.6.

Side by side

Grok-3 miniGrok 4.6
ProviderxAIxAI
Noometry Index41.056.8
Released2025-04-092026-08-12
WeightsProprietaryProprietary
Context window—500K
Max output—500K
Input $ / M tokens—$2
Output $ / M tokens—$6
Results tracked3549

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

Category by category

Coding Grok 4.6 leads

Grok-3 mini: 40.4 (#126), Grok 4.6: 58.0 (#14)

BenchmarkGrok-3 miniGrok 4.6
WeirdML42.6%67.3%
LMArena Coding13801464
DeepSWE—67.5%
FrontierCode—48%
Aider Polyglot49.3%—
CursorBench—41.4%
LMArena WebDev—1620
FrontierSWE—25.3%
SciCode—56.5%
ALE-Bench—1,508

Agentic & Tool Use Not comparable

Grok-3 mini: —, Grok 4.6: 39.4 (#25)

BenchmarkGrok-3 miniGrok 4.6
APEX-Agents—65.3%
GDP.pdf—17.2%
Vending-Bench 2—9,047

Reasoning Grok 4.6 leads

Grok-3 mini: 13.8 (#323), Grok 4.6: 61.3 (#18)

BenchmarkGrok-3 miniGrok 4.6
ARC-AGI-20.4%67.1%
ARC-AGI-116.5%87.5%
LMArena Hard Prompts13741445
Epoch Capabilities Index140.35156.48
SimpleBench—75.9%
Kagi LLM Benchmark61.3%—
NYT Connections (extended)—80%
CritPt—19.7%
Chess Puzzles—40%
EBR-Bench—30.5%
Mystery Game Puzzles—34%
DTBench—97.3%
LMCA—48.5%

Math Grok 4.6 leads

Grok-3 mini: 40.3 (#78), Grok 4.6: 66.8 (#19)

BenchmarkGrok-3 miniGrok 4.6
OTIS Mock AIME 2024-202577.8%99.2%
LMArena Math13861416
FrontierMath (Tiers 1-3)—66%
FrontierMath Tier 4—31.7%
ProofBench—51%
Omni-MATH31.8%—
MATH Level 590.9%—
FrontierMath (Feb 2025 set)5.9%—

Knowledge Grok 4.6 leads

Grok-3 mini: 46.3 (#75), Grok 4.6: 63.3 (#18)

BenchmarkGrok-3 miniGrok 4.6
GPQA Diamond76.3%94%
LMArena Expert13951463
SimpleQA Verified—49.3%
MMLU-Pro79.9%—
Confabulations10.8%—
GPQA (HELM)67.5%—

Multimodal Not comparable

Grok-3 mini: —, Grok 4.6: 43.9 (#18)

BenchmarkGrok-3 miniGrok 4.6
LMArena Vision—1265
Blueprint-Bench 2—33.2%
Furniture Assembly—40%
LMArena Document—1452

Multilingual Grok 4.6 leads

Grok-3 mini: 48.3 (#141), Grok 4.6: 53.3 (#67)

BenchmarkGrok-3 miniGrok 4.6
LMArena Non-English13521422
LMArena Chinese13871479
LMArena French13571458
LMArena German13501425
LMArena Japanese13431383
LMArena Korean13361395
LMArena Russian13521427
LMArena Spanish13791408

Instruction Following Grok-3 mini leads

Grok-3 mini: 78.7 (#8), Grok 4.6: 75.6 (#59)

BenchmarkGrok-3 miniGrok 4.6
LMArena Instruction Following13561431
IFEval95.1%—

Long Context Grok 4.6 leads

Grok-3 mini: 41.0 (#142), Grok 4.6: 44.4 (#62)

BenchmarkGrok-3 miniGrok 4.6
LMArena Longer Query13721453
Fiction.LiveBench66.7%—

Writing & Preference Grok 4.6 leads

Grok-3 mini: 52.7 (#163), Grok 4.6: 62.5 (#77)

BenchmarkGrok-3 miniGrok 4.6
LMArena Text13701427
LMArena Creative Writing13421430
LMArena Multi-Turn13551422
Short-Story Creative Writing73.5%—
WildBench65.1%—

Frequently asked questions

Is Grok-3 mini better than Grok 4.6?

Grok 4.6 is the stronger model overall, scoring 56.8 to 41.0 on the Noometry Index.

Is Grok-3 mini or Grok 4.6 better for coding?

Grok 4.6 scores higher on coding benchmarks: 58.0 versus 40.4 in the Noometry coding category.

How many benchmarks do Grok-3 mini and Grok 4.6 share?

23 benchmarks have published results for both models. Grok-3 mini has 35 scored results on Noometry and Grok 4.6 has 49.

Related comparisons

Go deeper