DataLearner logo

Kimi K2.6vsKimi K2.5

Across 12 shared benchmarks, Kimi K2.6 leads overall: Kimi K2.6 wins 11, Kimi K2.5 wins 1, with 0 ties and an average score difference of +6.27.

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 K2.611 wins(92%)(8%)1 winKimi K2.5

Benchmark scores

Grouped by capability, sorted by largest gap within each. 12 shared benchmarks.

Coding and Software Engineer

Kimi K2.6 4/4
BenchmarkKimi K2.6Kimi K2.5Diff
SWE-Bench Pro - Public58.6013 / 54Thinking (With Tools)50.7041 / 54Thinking (With Tools)+7.90
LiveCodeBench89.607 / 123Thinking (No Tools)8516 / 123Thinking (No Tools)+4.60
SWE-bench Multilingual76.705 / 23Thinking (With Tools)7313 / 23Thinking (No Tools)+3.70
SWE-bench Verified80.2014 / 112Thinking (With Tools)76.8030 / 112Thinking (With Tools)+3.40

General Knowledge

Kimi K2.6 3/3
BenchmarkKimi K2.6Kimi K2.5Diff
HLE5415 / 172Thinking (With Tools + Internet)50.2027 / 172Thinking (With Tools)+3.80
LiveBench72.1728 / 115Thinking (No Tools)69.0742 / 115Thinking (No Tools)+3.10
GPQA Diamond90.5018 / 187Thinking (No Tools)87.6037 / 187Thinking (No Tools)+2.90

Math and Reasoning

Kimi K2.6 2/2
BenchmarkKimi K2.6Kimi K2.5Diff
IMO-AnswerBench868 / 21Thinking (No Tools)81.8016 / 21Thinking (No Tools)+4.20
AIME 202696.403 / 18Thinking (No Tools)92.5012 / 18Thinking (No Tools)+3.90

AI Agent - Information Search

Kimi K2.6 1/1
BenchmarkKimi K2.6Kimi K2.5Diff
BrowseComp83.2014 / 53Thinking (With Tools + Internet)60.6036 / 53Thinking (With Tools + Internet)+22.60

AI Agent - Tool Usage

Kimi K2.6 1/1
BenchmarkKimi K2.6Kimi K2.5Diff
Terminal Bench 2.066.7010 / 47Thinking (With Tools)50.8034 / 47Thinking (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 tokens$0.6 / 1M tokens
Text output$4 / 1M tokens$3 / 1M tokens
Cache read$0.16 / 1M tokens$0.1 / 1M tokens
Cache write$0.95 / 1M tokensNot public

Summary

  • Kimi K2.6leads in:Coding and Software Engineer (4/4), General Knowledge (3/3), 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 12 shared benchmarks, Kimi K2.6 scores 6.27 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.