Crosshair
Leaderboard
Language ModelsOpen weights

Kimi K2.6

Moonshot's trillion-param MoE (32B active); strong agentic coding (Modified MIT).

textcodeOfficial site
Crosshair Index
70.5
#10 of 32 · Language Models
Provider
Moonshot AI
Released
2026-04-20
Parameters
1000B
Context
262.144K tokens

Token pricing

Input
$0.95 /1M
Output
$4 /1M
Cache read
$0.16 /1M

USD per 1M tokens · cache read = cached input (hit), cache write = caching surcharge · official list pricing (June 2026).

Industry skill web

Professional-domain strengths, composed from the benchmarks relevant to each field. Highlight an axis to see the benchmarks behind it.

SoftwareIB / FinanceLawMedicineResearchConsultingAccounting

Software Engineering

85
skill

Shipping working code against real repositories: bug fixes, feature patches, and competitive programming under tests.

SWE-bench Verified80.2%
LiveCodeBench89.6%

Scorecard

BenchmarkScoreSourceStatus
GPQA Diamond
Reasoning
90.5%Moonshot — Kimi K2.6 model cardvendorunverified
Humanity's Last Exam
Frontier
34.7%Moonshot — Kimi K2.6 model cardvendorunverified
SWE-bench Verified
Agentic Coding
80.2%Moonshot — Kimi K2.6 model cardvendorunverified
LiveCodeBench
Coding
89.6%Moonshot — Kimi K2.6 model cardvendorunverified
MMLU-Pro
Knowledge
—not evaluated
LMArena Elo
Human Preference
1,462LMArena (arena.ai)3rd-partyunverified
AA Intelligence Index
Composite
54Artificial Analysis3rd-partyunverified
Corporate Finance
Finance
66.7%Vals AI — CorpFin v23rd-partyunverified
LegalBench
Law
84.7%Vals AI — LegalBench3rd-partyunverified
TaxEval
Tax & Accounting
74.7%Vals AI — TaxEval v23rd-partyunverified
Medical Coding
Medicine
40.1%Vals AI — MedCode3rd-partyunverified