Noometry

Model comparison

Grok 3 vs Grok 4.6

Grok 4.6 is the stronger model overall, scoring 56.9 to 39.9 on the Noometry Index.

Last verified . 24 shared benchmarks.

Grok 3 xAI

39.9

Rank #157 Confirmed

Grok 4.6 xAI

56.9

Rank #21 Confirmed

Summary

  • They share 24 benchmarks with published results for both. Grok 3 scores higher in 0 categories and Grok 4.6 in 9 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Grok 4.6 leads 61.4 to 13.7.
  • The biggest single-benchmark swing is ARC-AGI-1: 5.5% for Grok 3 and 87.5% for Grok 4.6.

Side by side

Grok 3Grok 4.6
ProviderxAIxAI
Noometry Index39.956.9
Released2025-04-092026-08-12
WeightsProprietaryProprietary
Context window—500K
Max output—500K
Input $ / M tokens—$2
Output $ / M tokens—$6
Results tracked4049

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

Category by category

Coding Grok 4.6 leads

Grok 3: 41.9 (#115), Grok 4.6: 58.5 (#16)

BenchmarkGrok 3Grok 4.6
WeirdML37.2%67.3%
LMArena Coding14321465
DeepSWE—67.5%
FrontierCode—48%
Aider Polyglot53.3%—
CursorBench—41.4%
LMArena WebDev—1617
FrontierSWE—25.3%
SciCode—56.5%
ALE-Bench—1,508

Agentic & Tool Use Grok 4.6 leads

Grok 3: 30.5 (#76), Grok 4.6: 39.4 (#27)

BenchmarkGrok 3Grok 4.6
APEX-Agents—65.3%
BALROG29.5%—
GDP.pdf—17.2%
Vending-Bench 2—9,047

Reasoning Grok 4.6 leads

Grok 3: 13.7 (#333), Grok 4.6: 61.4 (#20)

BenchmarkGrok 3Grok 4.6
ARC-AGI-20%67.1%
SimpleBench36.1%75.9%
ARC-AGI-15.5%87.5%
LMArena Hard Prompts14341447
Epoch Capabilities Index138.33156.44
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: 38.0 (#145), Grok 4.6: 67.0 (#24)

BenchmarkGrok 3Grok 4.6
OTIS Mock AIME 2024-202555.6%99.2%
LMArena Math13911423
FrontierMath (Tiers 1-3)—66%
FrontierMath Tier 4—31.7%
ProofBench—51%
Omni-MATH46.4%—
MATH Level 588.7%—
FrontierMath (Feb 2025 set)3.8%—
FrontierMath Tier 4 (v1)0%—

Knowledge Grok 4.6 leads

Grok 3: 46.2 (#82), Grok 4.6: 63.3 (#20)

BenchmarkGrok 3Grok 4.6
GPQA Diamond75.8%94%
LMArena Expert14211467
SimpleQA Verified—49.3%
MMLU-Pro78.8%—
Confabulations14.2%—
Vectara Hallucination Rate5.8%—
GPQA (HELM)65%—

Multimodal Not comparable

Grok 3: —, Grok 4.6: 43.6 (#23)

BenchmarkGrok 3Grok 4.6
LMArena Vision—1263
Blueprint-Bench 2—33.2%
Furniture Assembly—40%
LMArena Document—1452

Multilingual Too close to call

Grok 3: 52.3 (#87), Grok 4.6: 53.0 (#74)

BenchmarkGrok 3Grok 4.6
LMArena Non-English14101420
LMArena Chinese14481480
LMArena French14601461
LMArena German14311431
LMArena Japanese13871376
LMArena Korean13731397
LMArena Russian14161422
LMArena Spanish14171404

Instruction Following Too close to call

Grok 3: 75.0 (#73), Grok 4.6: 75.4 (#63)

BenchmarkGrok 3Grok 4.6
LMArena Instruction Following14091431
IFEval88.4%—

Long Context Grok 4.6 leads

Grok 3: 38.7 (#192), Grok 4.6: 44.5 (#66)

BenchmarkGrok 3Grok 4.6
LMArena Longer Query14391454
Fiction.LiveBench58.3%—

Writing & Preference Grok 4.6 leads

Grok 3: 55.8 (#141), Grok 4.6: 62.3 (#80)

BenchmarkGrok 3Grok 4.6
LMArena Text14261428
LMArena Creative Writing14141428
LMArena Multi-Turn14251425
Short-Story Creative Writing76.4%—
EQ-Bench Creative Writing1186—
WildBench84.9%—

Frequently asked questions

Is Grok 3 better than Grok 4.6?

Grok 4.6 is the stronger model overall, scoring 56.9 to 39.9 on the Noometry Index.

Is Grok 3 or Grok 4.6 better for coding?

Grok 4.6 scores higher on coding benchmarks: 58.5 versus 41.9 in the Noometry coding category.

How many benchmarks do Grok 3 and Grok 4.6 share?

24 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and Grok 4.6 has 49.

Related comparisons

Go deeper