DataLearner logo

Kimi K2.7 CodevsKimi K2.6

Across 7 shared benchmarks, Kimi K2.7 Code leads overall: Kimi K2.7 Code wins 6, Kimi K2.6 wins 1, with 0 ties and an average score difference of +7.56.

Moonshot AI
Kimi K2.7 Code

Moonshot AI · 2026-06-12 · Coding model

Moonshot AI
Kimi K2.6

Moonshot AI · 2026-04-20 · Reasoning model

Kimi K2.7 Code6 wins(86%)(14%)1 winKimi K2.6

Benchmark scores

Grouped by capability, sorted by largest gap within each. 7 shared benchmarks.

AI Agent - Tool Usage

Kimi K2.7 Code 3/3
BenchmarkKimi K2.7 CodeKimi K2.6Diff
Terminal-Bench 2.167.0433 / 43Thinking (With Tools)53.5642 / 43Thinking (No Tools)+13.48
MCPMark-Verified81.102 / 3Thinking (With Tools)72.803 / 3Thinking (With Tools)+8.30
MCP-Atlas7617 / 38Thinking (With Tools)69.4028 / 38Thinking (With Tools)+6.60

Coding and Software Engineer

Kimi K2.7 Code 3/3
BenchmarkKimi K2.7 CodeKimi K2.6Diff
Kimi Code Bench 2.0622 / 3Thinking (With Tools)50.903 / 3Thinking (With Tools)+11.10
MLS Bench35.103 / 4Thinking (With Tools)26.704 / 4Thinking (With Tools)+8.40
Program Bench53.603 / 5Thinking (With Tools)48.304 / 5Thinking (With Tools)+5.30

General Knowledge

Kimi K2.6 1/1
BenchmarkKimi K2.7 CodeKimi K2.6Diff
LiveBench71.8930 / 115Normal (No Tools)72.1728 / 115Thinking (No Tools)-0.28

Specs

FieldKimi K2.7 CodeKimi K2.6
PublisherMoonshot AIMoonshot AI
Release date2026-06-122026-04-20
Model typeCoding modelReasoning model
ArchitectureMoEMoE
Parameters1T1T
Context length256K256K
Max outputNot availableNot available

API pricing

Prices use DataLearner records when available; missing fields are not inferred.

ItemKimi K2.7 CodeKimi K2.6
Text input$0.95 / 1M tokens$0.95 / 1M tokens
Text output$4 / 1M tokens$4 / 1M tokens
Cache read$0.19 / 1M tokens$0.16 / 1M tokens
Cache writeNot public$0.95 / 1M tokens

Summary

  • Kimi K2.7 Codeleads in:AI Agent - Tool Usage (3/3), Coding and Software Engineer (3/3)
  • Kimi K2.6leads in:General Knowledge (1/1)

On average across the 7 shared benchmarks, Kimi K2.7 Code scores 7.56 higher.

Largest single-benchmark gap: Terminal-Bench 2.1 — Kimi K2.7 Code 67.04 vs Kimi K2.6 53.56 (+13.48).

Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.