Kimi K2.6vsKimi K2.5

Across 11 shared benchmarks, Kimi K2.6 leads overall: Kimi K2.6 wins 10, Kimi K2.5 wins 1, with 0 ties and an average score difference of +6.55.

Moonshot AI
Kimi K2.6

Moonshot AI · 2026-04-20 · Reasoning model

Moonshot AI
Kimi K2.5

Moonshot AI · 2026-01-27 · Multimodal model

Kimi K2.610 wins(91%)(9%)1 winKimi K2.5

Benchmark scores

Grouped by capability, sorted by largest gap within each. 11 shared benchmarks.

Coding and Software Engineer

Kimi K2.6 4/4
BenchmarkKimi K2.6Kimi K2.5Diff
SWE-Bench Pro - Public58.607 / 43Thinking (With Tools)50.7031 / 43Thinking (With Tools)+7.90
LiveCodeBench89.607 / 120Thinking (No Tools)8516 / 120Thinking (No Tools)+4.60
SWE-bench Multilingual76.704 / 20Thinking (With Tools)7311 / 20Thinking (No Tools)+3.70
SWE-bench Verified80.2013 / 108Thinking (With Tools)76.8027 / 108Thinking (With Tools)+3.40

General Knowledge

Kimi K2.6 2/2
BenchmarkKimi K2.6Kimi K2.5Diff
HLE549 / 157Thinking (With Tools + Internet)50.2020 / 157Thinking (With Tools)+3.80
GPQA Diamond90.5016 / 178Thinking (No Tools)87.6034 / 178Thinking (No Tools)+2.90

Math and Reasoning

Kimi K2.6 2/2
BenchmarkKimi K2.6Kimi K2.5Diff
IMO-AnswerBench866 / 19Thinking (No Tools)81.8014 / 19Thinking (No Tools)+4.20
AIME 202696.401 / 14Thinking (No Tools)92.5010 / 14Thinking (No Tools)+3.90

AI Agent - Information Search

Kimi K2.6 1/1
BenchmarkKimi K2.6Kimi K2.5Diff
BrowseComp83.2010 / 45Thinking (With Tools + Internet)60.6029 / 45Thinking (With Tools + Internet)+22.60

AI Agent - Tool Usage

Kimi K2.6 1/1
BenchmarkKimi K2.6Kimi K2.5Diff
Terminal Bench 2.066.7010 / 46Thinking (With Tools)50.8033 / 46Thinking (With Tools)+15.90

Claw-style Agent Evaluation

Kimi K2.5 1/1
BenchmarkKimi K2.6Kimi K2.5Diff
Claw Bench80.9019 / 29Thinking (With Tools)81.7018 / 29Thinking (With Tools)-0.80

Specs

FieldKimi K2.6Kimi K2.5
PublisherMoonshot AIMoonshot AI
Release date2026-04-202026-01-27
Model typeReasoning modelMultimodal model
ArchitectureMoEMoE
Parameters1T1T
Context length256K256K
Max outputNot available16K

API pricing

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

ItemKimi K2.6Kimi K2.5
Text input$0.95 / 1M tokensNot public
Text output$4 / 1M tokensNot public
Cache read$0.16 / 1M tokensNot public
Cache write$0.95 / 1M tokensNot public

One or both models have incomplete public pricing.

Summary

  • Kimi K2.6leads in:Coding and Software Engineer (4/4), General Knowledge (2/2), Math and Reasoning (2/2), AI Agent - Information Search (1/1), AI Agent - Tool Usage (1/1)
  • Kimi K2.5leads in:Claw-style Agent Evaluation (1/1)

On average across the 11 shared benchmarks, Kimi K2.6 scores 6.55 higher.

Largest single-benchmark gap: BrowseComp — Kimi K2.6 83.20 vs Kimi K2.5 60.60 (+22.60).

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