Noometry

Model comparison

Grok 3 vs Kimi K3

Kimi K3 is the stronger model overall, scoring 59.5 to 39.9 on the Noometry Index.

Last verified . 25 shared benchmarks.

Grok 3 xAI

39.9

Rank #157 Confirmed

Kimi K3 Moonshot AI

59.5

Rank #15 Confirmed

Summary

  • They share 25 benchmarks with published results for both. Grok 3 scores higher in 0 categories and Kimi K3 in 9 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Kimi K3 leads 63.0 to 13.7.
  • The biggest single-benchmark swing is ARC-AGI-1: 5.5% for Grok 3 and 94.5% for Kimi K3.
  • Kimi K3 has downloadable open weights; the other is API-only.

Side by side

Grok 3Kimi K3
ProviderxAIMoonshot AI
Noometry Index39.959.5
Released2025-04-092026-07-16
WeightsProprietaryOpen
Context window—1.05M
Max output—1.05M
Input $ / M tokens—$3
Output $ / M tokens—$15
Results tracked4053

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

Category by category

Coding Kimi K3 leads

Grok 3: 41.9 (#115), Kimi K3: 61.0 (#10)

BenchmarkGrok 3Kimi K3
WeirdML37.2%82.6%
LMArena Coding14321508
DeepSWE—68.5%
FrontierCode—44.2%
Aider Polyglot53.3%—
LMArena WebDev—1654
FrontierSWE—25.9%
SciCode—59.5%
ALE-Bench—1,524

Agentic & Tool Use Kimi K3 leads

Grok 3: 30.5 (#76), Kimi K3: 41.8 (#20)

BenchmarkGrok 3Kimi K3
APEX-Agents—50.6%
τ²-bench Banking—37.1%
PostTrainBench—32%
BALROG29.5%—
GBAEval—48.3%
GDP.pdf—19%
Vending-Bench 2—5,165

Reasoning Kimi K3 leads

Grok 3: 13.7 (#333), Kimi K3: 63.0 (#17)

BenchmarkGrok 3Kimi K3
ARC-AGI-20%60.4%
SimpleBench36.1%60.7%
ARC-AGI-15.5%94.5%
LMArena Hard Prompts14341496
Epoch Capabilities Index138.33157.45
Kagi LLM Benchmark61.3%—
NYT Connections (extended)—93.6%
CritPt—23.4%
Chess Puzzles—39%
Mystery Game Puzzles—26%
DTBench—91.2%
LMCA—52.7%
Surface Evolver Bench—95%
ForecastBench—61.1

Math Kimi K3 leads

Grok 3: 38.0 (#145), Kimi K3: 74.2 (#16)

BenchmarkGrok 3Kimi K3
OTIS Mock AIME 2024-202555.6%97.2%
LMArena Math13911491
FrontierMath (Tiers 1-3)—72.2%
FrontierMath Tier 4—39%
MathArena Final-Answer Competitions—87.8%
ProofBench—87%
Omni-MATH46.4%—
MATH Level 588.7%—
FrontierMath (Feb 2025 set)3.8%—
FrontierMath Tier 4 (v1)0%—

Knowledge Kimi K3 leads

Grok 3: 46.2 (#82), Kimi K3: 63.2 (#21)

BenchmarkGrok 3Kimi K3
GPQA Diamond75.8%93.1%
LMArena Expert14211521
SimpleQA Verified—50.6%
MMLU-Pro78.8%—
Confabulations14.2%—
Vectara Hallucination Rate5.8%—
GPQA (HELM)65%—

Multimodal Not comparable

Grok 3: —, Kimi K3: 37.8 (#70)

BenchmarkGrok 3Kimi K3
Blueprint-Bench 2—29.5%
Furniture Assembly—34.2%

Multilingual Kimi K3 leads

Grok 3: 52.3 (#87), Kimi K3: 56.3 (#21)

BenchmarkGrok 3Kimi K3
LMArena Non-English14101466
LMArena Chinese14481529
LMArena French14601491
LMArena German14311488
LMArena Japanese13871487
LMArena Korean13731458
LMArena Russian14161482
LMArena Spanish14171472

Instruction Following Kimi K3 leads

Grok 3: 75.0 (#73), Kimi K3: 77.7 (#14)

BenchmarkGrok 3Kimi K3
LMArena Instruction Following14091483
IFEval88.4%—

Long Context Kimi K3 leads

Grok 3: 38.7 (#192), Kimi K3: 45.8 (#29)

BenchmarkGrok 3Kimi K3
LMArena Longer Query14391494
Fiction.LiveBench58.3%—

Writing & Preference Kimi K3 leads

Grok 3: 55.8 (#141), Kimi K3: 76.6 (#4)

BenchmarkGrok 3Kimi K3
LMArena Text14261476
LMArena Creative Writing14141454
EQ-Bench Creative Writing11862082
LMArena Multi-Turn14251488
Short-Story Creative Writing76.4%—
WildBench84.9%—
EQ-Bench 4—1339

Frequently asked questions

Is Grok 3 better than Kimi K3?

Kimi K3 is the stronger model overall, scoring 59.5 to 39.9 on the Noometry Index.

Is Grok 3 or Kimi K3 better for coding?

Kimi K3 scores higher on coding benchmarks: 61.0 versus 41.9 in the Noometry coding category.

How many benchmarks do Grok 3 and Kimi K3 share?

25 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and Kimi K3 has 53.

Related comparisons

Go deeper