Noometry

Model comparison

Grok 4 vs Kimi K3

Kimi K3 is the stronger model overall, scoring 59.5 to 48.1 on the Noometry Index.

Last verified . 26 shared benchmarks.

Grok 4 xAI

48.1

Rank #56 Confirmed

Kimi K3 Moonshot AI

59.5

Rank #15 Confirmed

Summary

  • They share 26 benchmarks with published results for both. Grok 4 scores higher in 2 categories and Kimi K3 in 8 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Kimi K3 leads 63.0 to 36.7.
  • The biggest single-benchmark swing is ARC-AGI-2: 16% for Grok 4 and 60.4% for Kimi K3.
  • Kimi K3 has downloadable open weights; the other is API-only.

Side by side

Grok 4Kimi K3
ProviderxAIMoonshot AI
Noometry Index48.159.5
Released2025-07-092026-07-16
WeightsProprietaryOpen
Context window—1.05M
Max output—1.05M
Input $ / M tokens—$3
Output $ / M tokens—$15
Results tracked4853

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

Category by category

Coding Kimi K3 leads

Grok 4: 50.3 (#46), Kimi K3: 61.0 (#10)

BenchmarkGrok 4Kimi K3
WeirdML45.7%82.6%
LMArena Coding14081508
DeepSWE—68.5%
FrontierCode—44.2%
Aider Polyglot79.6%—
LMArena WebDev—1654
FrontierSWE—25.9%
SciCode—59.5%
ALE-Bench—1,524

Agentic & Tool Use Kimi K3 leads

Grok 4: 32.3 (#68), Kimi K3: 41.8 (#20)

BenchmarkGrok 4Kimi K3
Terminal-Bench27.2%—
APEX-Agents—50.6%
Berkeley Function Calling Leaderboard63%—
GDPval21.1%—
τ²-bench Banking—37.1%
Cybench43%—
DeepResearch Bench47.3%—
PostTrainBench—32%
BALROG43.6%—
GBAEval—48.3%
GDP.pdf—19%
LMArena Search1142—
METR Time Horizons66.6%—
Vending-Bench 2—5,165

Reasoning Kimi K3 leads

Grok 4: 36.7 (#65), Kimi K3: 63.0 (#17)

BenchmarkGrok 4Kimi K3
ARC-AGI-216%60.4%
SimpleBench60.5%60.7%
ARC-AGI-166.7%94.5%
Chess Puzzles28%39%
LMArena Hard Prompts14091496
Epoch Capabilities Index146.44157.45
ForecastBench60.961.1
Kagi LLM Benchmark73.6%—
NYT Connections (extended)—93.6%
CritPt—23.4%
Mystery Game Puzzles—26%
DTBench—91.2%
LMCA—52.7%
Surface Evolver Bench—95%

Math Kimi K3 leads

Grok 4: 48.4 (#64), Kimi K3: 74.2 (#16)

BenchmarkGrok 4Kimi K3
OTIS Mock AIME 2024-202584%97.2%
LMArena Math14221491
FrontierMath (Tiers 1-3)—72.2%
FrontierMath Tier 4—39%
MathArena Final-Answer Competitions—87.8%
ProofBench—87%
Omni-MATH60.3%—
FrontierMath (Feb 2025 set)19.7%—
FrontierMath Tier 4 (v1)2.1%—

Knowledge Kimi K3 leads

Grok 4: 53.8 (#55), Kimi K3: 63.2 (#21)

BenchmarkGrok 4Kimi K3
GPQA Diamond87%93.1%
LMArena Expert14151521
SimpleQA Verified—50.6%
MMLU-Pro85.1%—
Confabulations12.4%—
GPQA (HELM)72.7%—

Multimodal Kimi K3 leads

Grok 4: 33.7 (#94), Kimi K3: 37.8 (#70)

BenchmarkGrok 4Kimi K3
LMArena Vision1210—
GeoBench45%—
Blueprint-Bench 2—29.5%
Furniture Assembly—34.2%

Multilingual Kimi K3 leads

Grok 4: 51.8 (#103), Kimi K3: 56.3 (#21)

BenchmarkGrok 4Kimi K3
LMArena Non-English14031466
LMArena Chinese14271529
LMArena French14181491
LMArena German14291488
LMArena Japanese13941487
LMArena Korean13771458
LMArena Russian14101482
LMArena Spanish14201472

Instruction Following Grok 4 leads

Grok 4: 79.2 (#5), Kimi K3: 77.7 (#14)

BenchmarkGrok 4Kimi K3
LMArena Instruction Following13871483
IFEval94.9%—

Long Context Grok 4 leads

Grok 4: 63.1 (#4), Kimi K3: 45.8 (#29)

BenchmarkGrok 4Kimi K3
LMArena Longer Query14091494
Fiction.LiveBench94.4%—

Writing & Preference Kimi K3 leads

Grok 4: 58.5 (#116), Kimi K3: 76.6 (#4)

BenchmarkGrok 4Kimi K3
LMArena Text14111476
LMArena Creative Writing13971454
LMArena Multi-Turn14161488
Short-Story Creative Writing76.9%—
EQ-Bench Creative Writing—2082
WildBench79.7%—
EQ-Bench 4—1339

Frequently asked questions

Is Grok 4 better than Kimi K3?

Kimi K3 is the stronger model overall, scoring 59.5 to 48.1 on the Noometry Index.

Is Grok 4 or Kimi K3 better for coding?

Kimi K3 scores higher on coding benchmarks: 61.0 versus 50.3 in the Noometry coding category.

How many benchmarks do Grok 4 and Kimi K3 share?

26 benchmarks have published results for both models. Grok 4 has 48 scored results on Noometry and Kimi K3 has 53.

Related comparisons

Go deeper