Noometry

Model comparison

Grok 3 vs Kimi K2.7 Code

Kimi K2.7 Code is the stronger model overall, scoring 43.0 to 39.6 on the Noometry Index.

Last verified . 5 shared benchmarks.

Grok 3 xAI

39.6

Rank #150 Confirmed

Kimi K2.7 Code Moonshot AI

43.0

Rank #90 Confirmed

Summary

  • They share 5 benchmarks with published results for both. Grok 3 scores higher in 1 category and Kimi K2.7 Code in 4 categories; 4 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Kimi K2.7 Code leads 38.2 to 13.8.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 55.6% for Grok 3 and 95.6% for Kimi K2.7 Code.
  • Kimi K2.7 Code has downloadable open weights; the other is API-only.

Side by side

Grok 3Kimi K2.7 Code
ProviderxAIMoonshot AI
Noometry Index39.643.0
Released2025-04-092026-06-12
WeightsProprietaryOpen
Context window—262K
Max output—262K
Input $ / M tokens—$0.95
Output $ / M tokens—$4
Results tracked4019

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

Category by category

Coding Too close to call

Grok 3: 41.5 (#113), Kimi K2.7 Code: 42.4 (#92)

BenchmarkGrok 3Kimi K2.7 Code
WeirdML37.2%54.1%
DeepSWE—30.5%
FrontierCode—30.1%
Aider Polyglot53.3%—
LMArena WebDev—1473
SciCode—47.5%
LMArena Coding1432—
ALE-Bench—886.23

Agentic & Tool Use Grok 3 leads

Grok 3: 30.2 (#69), Kimi K2.7 Code: 23.9 (#115)

BenchmarkGrok 3Kimi K2.7 Code
APEX-Agents—37.6%
BALROG29.5%—
GBAEval—0.9%
Vending-Bench 2—5,083

Reasoning Kimi K2.7 Code leads

Grok 3: 13.8 (#322), Kimi K2.7 Code: 38.2 (#56)

BenchmarkGrok 3Kimi K2.7 Code
SimpleBench36.1%57.9%
Epoch Capabilities Index138.34150.01
ARC-AGI-20%—
Kagi LLM Benchmark61.3%—
ARC-AGI-15.5%—
CritPt—10%
Chess Puzzles—21%
LMArena Hard Prompts1433—
Surface Evolver Bench—48.8%

Math Kimi K2.7 Code leads

Grok 3: 36.2 (#122), Kimi K2.7 Code: 52.6 (#39)

BenchmarkGrok 3Kimi K2.7 Code
OTIS Mock AIME 2024-202555.6%95.6%
FrontierMath (Tiers 1-3)—54%
FrontierMath Tier 4—12.2%
Omni-MATH46.4%—
LMArena Math1390—
MATH Level 588.7%—
FrontierMath (Feb 2025 set)3.8%—
FrontierMath Tier 4 (v1)0%—

Knowledge Kimi K2.7 Code leads

Grok 3: 46.0 (#76), Kimi K2.7 Code: 53.6 (#51)

BenchmarkGrok 3Kimi K2.7 Code
GPQA Diamond75.8%87.9%
SimpleQA Verified—36.5%
MMLU-Pro78.8%—
Confabulations14.2%—
Vectara Hallucination Rate5.8%—
GPQA (HELM)65%—
LMArena Expert1420—

Multilingual Not comparable

Grok 3: 52.5 (#81), Kimi K2.7 Code: —

BenchmarkGrok 3Kimi K2.7 Code
LMArena Non-English1411—
LMArena Chinese1448—
LMArena French1460—
LMArena German1431—
LMArena Japanese1389—
LMArena Korean1374—
LMArena Russian1416—
LMArena Spanish1417—

Instruction Following Not comparable

Grok 3: 75.2 (#69), Kimi K2.7 Code: —

BenchmarkGrok 3Kimi K2.7 Code
IFEval88.4%—
LMArena Instruction Following1409—

Long Context Not comparable

Grok 3: 38.7 (#187), Kimi K2.7 Code: —

BenchmarkGrok 3Kimi K2.7 Code
Fiction.LiveBench58.3%—
LMArena Longer Query1439—

Writing & Preference Not comparable

Grok 3: 56.0 (#136), Kimi K2.7 Code: —

BenchmarkGrok 3Kimi K2.7 Code
LMArena Text1426—
LMArena Creative Writing1414—
Short-Story Creative Writing76.4%—
EQ-Bench Creative Writing1186—
WildBench84.9%—
LMArena Multi-Turn1424—

Frequently asked questions

Is Grok 3 better than Kimi K2.7 Code?

Kimi K2.7 Code is the stronger model overall, scoring 43.0 to 39.6 on the Noometry Index.

Is Grok 3 or Kimi K2.7 Code better for coding?

They score almost the same on coding (41.5 vs 42.4); test both on your own repository before choosing.

How many benchmarks do Grok 3 and Kimi K2.7 Code share?

5 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and Kimi K2.7 Code has 19.

Related comparisons

Go deeper