Kimi K3 (Max)
KimiKimi오픈 웨이트Kimi K3 License · 상업적 사용 가능
설명
Kimi K3 is Moonshot AI's flagship open Mixture-of-Experts model for long-horizon coding, knowledge work, and reasoning. It has 2.8 trillion total parameters, activates 16 of 896 experts, and combines Kimi Delta Attention, Attention Residuals, and Stable LatentMoE. The model supports a 1M-token context window, native image and video understanding, tool calling, structured output, and always-on reasoning with max thinking effort at launch.
출시일
2026-07-16
파라미터
2.8T
컨텍스트 길이
1.0M
모달리티
image, text, video
능력 레이더
44
general
74
coding
93
reasoning
72
science
70
agents
80
multimodal
랭킹
벤치마크 점수 (LLM Stats)
(LLM Stats (zeroeval))Agents
AA-Briefcase
1548.00 / 3000
Program Bench
77.8%자체 보고
DECK-Bench
73.5%자체 보고
Toolathlon
73.2%자체 보고
Kimi Code Bench v2
72.9%자체 보고
OfficeQA Pro
63.3%자체 보고
Job Bench
52.9%자체 보고
MLS-Bench Lite
48.3%자체 보고
PostTrainBench
36.6%자체 보고
SpreadsheetBench 2
34.8%자체 보고
AutomationBench
30.8%자체 보고
Code
FrontierSWE
81.2%자체 보고
DeepSWE 1.1
69.0%
DeepSWE
67.5%자체 보고
SWE-Marathon
42.0%자체 보고
Math
MathVision
97.8%자체 보고
Reasoning
GPQANYU + Cohere + Anthropic (2023)
93.5%자체 보고
CharXiv-R
91.3%자체 보고
BrowseCompOpenAI (2025)
91.2%자체 보고
Terminal-Bench 2.1
88.3%자체 보고
MCP Atlas
84.2%자체 보고
Humanity's Last Exam
56.0%자체 보고
APEX-Agents
37.6%자체 보고
Search
DeepSearchQA
95.0%자체 보고
Vision
OmniDocBench
91.1%자체 보고
BabyVision
85.7%자체 보고
MMMU-Pro (with tools)
83.4%자체 보고
MMMU-Pro
81.6%자체 보고
PerceptionBench
58.5%자체 보고
WorldVQA
51.0%자체 보고
ZEROBench
0.41 / 100자체 보고
AA 평가 지수
(Artificial Analysis)Gpqa(NYU + Cohere + Anthropic (2023))93.5
Lcr(Artificial Analysis)88.7
Terminalbench V2 185.0
Coding Index(Artificial Analysis)76.2
Scicode(UIUC + Argonne National Lab (2024))59.5
Hle(Center for AI Safety + Scale AI (2025))46.9
Tau Banking46.0
Intelligence Index(Artificial Analysis)43.6
Terminalbench V4 012.6
LLM Stats 카테고리 점수
(LLM Stats (zeroeval))Productivity100
Agents92
Reasoning78
General60
Physics90
Search90
Biology90
Chemistry90
Math80
Multimodal80
Code70
Tool Calling70
Vision70
Systems40
가격
입력 가격$3 / 1M 토큰
출력 가격$15 / 1M 토큰
혼합 가격 (3:1)$6 / 1M 토큰
캐시 읽기 가격$0.169 / 1M 토큰
속도
토큰/초0.0
첫 토큰 지연0.00s
첫 응답 지연0.00s
공급자 가격 순위
공급자 가격 순위
34개 공급자
최저가: DeepInfra최고가: Tinfoil
공급자입력출력
1DeepInfra최저가
$0
$0.00001
2Together
$0
$0.00002
3Moonshot AI
$0
$0.00002
4Fireworks
$0
$0.00002
5Novita
$0
$0.00002
6Kenari
$0.85714
$8.57143
7engy
$1.95
$9.75
8Boundless
$2.3
$11.4
9Vancine
$2.4
$12
10Alibaba (China)
$2.827
$14.133
11Moonshot AI (China)
$2.9879
$14.9395
12Kimi주요
$3
$15
13302.AI
$3
$15
14Alibaba
$3
$15
15Moark
$3
$15
16OpenCode Go
$3
$15
17Vivgrid
$3
$15
18GitHub Copilot
$3
$15
19OpenCode
$3
$15
20Ollama Cloud
$3
$15
21AIHubMix
$3
$15
22Requesty
$3
$15
23DigitalOcean
$3
$15
24Cortecs
$3
$14.999
25DevPass (LLM Gateway)
$3
$15
26Neuralwatt
$3
$15
27Neon
$3
$15
28EmpirioLabs AI
$3
$15
29Opper
$3
$15
30Melious
$3.19628
$15.98138
31Charm Hyper
$3.2664
$16.332
32Venice AI
$3.75
$18.75
33GreenPT
$3.762
$18.81
34Tinfoil
$4
$20
이 모델의 다양한 API 공급자 간 가격 비교.