Noometry

Model comparison

Kimi K2.6 vs Kimi K2.7 Code

Kimi K2.6 is the stronger model overall, scoring 47.7 to 43.3 on the Noometry Index.

Last verified . 14 shared benchmarks.

Kimi K2.6 Moonshot AI

47.7

Rank #60 Confirmed

Kimi K2.7 Code Moonshot AI

43.3

Rank #94 Confirmed

Summary

  • They share 14 benchmarks with published results for both. Kimi K2.6 scores higher in 4 categories and Kimi K2.7 Code in 1 category; 4 gaps are clear of the uncertainty.
  • The widest gap is in coding, where Kimi K2.6 leads 50.7 to 42.9.
  • The biggest single-benchmark swing is FrontierMath Tier 4: 25.6% for Kimi K2.6 and 12.2% for Kimi K2.7 Code.
  • Both cost about the same: $0.95 input and $4 output per million tokens.

Side by side

Kimi K2.6Kimi K2.7 Code
ProviderMoonshot AIMoonshot AI
Noometry Index47.743.3
Released2026-04-202026-06-12
WeightsOpenOpen
Context window262K262K
Max output262K262K
Input $ / M tokens$0.95$0.95
Output $ / M tokens$4$4
Results tracked5119

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

Category by category

Coding Kimi K2.6 leads

Kimi K2.6: 50.7 (#43), Kimi K2.7 Code: 42.9 (#95)

BenchmarkKimi K2.6Kimi K2.7 Code
LMArena WebDev15091473
SciCode53.5%47.5%
WeirdML55.9%54.1%
ALE-Bench1,093886.23
SWE-bench Verified76.7%—
DeepSWE—30.5%
FrontierCode—30.1%
LMArena Coding1488—

Agentic & Tool Use Kimi K2.7 Code leads

Kimi K2.6: 21.9 (#137), Kimi K2.7 Code: 24.0 (#122)

BenchmarkKimi K2.6Kimi K2.7 Code
GBAEval0.9%0.9%
Vending-Bench 26,2055,083
APEX-Agents—37.6%
OSWorld 2.04.6%—
ExploitBench18.4%—
GDP.pdf12%—

Reasoning Kimi K2.6 leads

Kimi K2.6: 40.5 (#55), Kimi K2.7 Code: 39.0 (#61)

BenchmarkKimi K2.6Kimi K2.7 Code
CritPt8%10%
Chess Puzzles26%21%
Epoch Capabilities Index151.05149.97
SimpleBench—57.9%
NYT Connections (extended)87.2%—
EBR-Bench2.4%—
LMArena Hard Prompts1470—
Mystery Game Puzzles18%—
DTBench90.9%—
LMCA37.3%—
Surface Evolver Bench—48.8%

Math Kimi K2.6 leads

Kimi K2.6: 57.0 (#41), Kimi K2.7 Code: 52.9 (#48)

BenchmarkKimi K2.6Kimi K2.7 Code
FrontierMath (Tiers 1-3)57.2%54%
FrontierMath Tier 425.6%12.2%
OTIS Mock AIME 2024-202596.1%95.6%
MathArena Final-Answer Competitions72.9%—
ProofBench16%—
LMArena Math1475—
FrontierMath (Feb 2025 set)39%—
FrontierMath Tier 4 (v1)14.6%—

Knowledge Too close to call

Kimi K2.6: 54.0 (#54), Kimi K2.7 Code: 53.5 (#57)

BenchmarkKimi K2.6Kimi K2.7 Code
GPQA Diamond90.8%87.9%
SimpleQA Verified34.9%36.5%
Vectara Hallucination Rate10.8%—
LMArena Expert1491—

Multimodal Not comparable

Kimi K2.6: 31.6 (#103), Kimi K2.7 Code: —

BenchmarkKimi K2.6Kimi K2.7 Code
LMArena Vision1283—
Blueprint-Bench 23.9%—
Furniture Assembly21.7%—
LMArena Document1451—

Multilingual Not comparable

Kimi K2.6: 54.9 (#37), Kimi K2.7 Code: —

BenchmarkKimi K2.6Kimi K2.7 Code
LMArena Non-English1446—
LMArena Chinese1521—
LMArena French1471—
LMArena German1450—
LMArena Japanese1443—
LMArena Korean1427—
LMArena Russian1446—
LMArena Spanish1464—

Instruction Following Not comparable

Kimi K2.6: 76.3 (#43), Kimi K2.7 Code: —

BenchmarkKimi K2.6Kimi K2.7 Code
LMArena Instruction Following1451—

Long Context Not comparable

Kimi K2.6: 44.9 (#52), Kimi K2.7 Code: —

BenchmarkKimi K2.6Kimi K2.7 Code
LMArena Longer Query1468—

Writing & Preference Not comparable

Kimi K2.6: 68.5 (#26), Kimi K2.7 Code: —

BenchmarkKimi K2.6Kimi K2.7 Code
LMArena Text1455—
LMArena Creative Writing1434—
EQ-Bench Creative Writing1725—
EQ-Bench 41202—
LMArena Multi-Turn1453—

Frequently asked questions

Is Kimi K2.6 better than Kimi K2.7 Code?

Kimi K2.6 is the stronger model overall, scoring 47.7 to 43.3 on the Noometry Index.

Which is cheaper, Kimi K2.6 or Kimi K2.7 Code?

Kimi K2.7 Code is cheaper. It lists at $0.95 per million input tokens and $4 per million output tokens; Kimi K2.6 lists at $0.95 and $4.

Is Kimi K2.6 or Kimi K2.7 Code better for coding?

Kimi K2.6 scores higher on coding benchmarks: 50.7 versus 42.9 in the Noometry coding category.

Which has the bigger context window?

Both accept 262K tokens.

How many benchmarks do Kimi K2.6 and Kimi K2.7 Code share?

14 benchmarks have published results for both models. Kimi K2.6 has 51 scored results on Noometry and Kimi K2.7 Code has 19.

Related comparisons

Go deeper