Kimi:模型智能、性能与价格

Kimi
Kimi

本分析旨在帮助你根据使用场景,选择 Kimi 提供的最佳模型。

最智能

Updated
#1
Kimi K3 (max)Kimi K3 (max)
44
#2
Kimi K3 (low)Kimi K3 (low)
30
#3
Kimi K2.6Kimi K2.6
27
#4
Kimi K2.7 CodeKimi K2.7 Code
26
#5
Kimi K2.6 (non-reasoning)Kimi K2.6 (non-reasoning)
24

Intelligence Index

共 5 个模型

速度最快

#1
Kimi K2.7 CodeKimi K2.7 Code
105 t/s
#2
Kimi K2.6Kimi K2.6
71 t/s
#3
Kimi K2.6 (non-reasoning)Kimi K2.6 (non-reasoning)
64 t/s
#4
Kimi K3 (max)Kimi K3 (max)
46 t/s
#5
Kimi K3 (low)Kimi K3 (low)
46 t/s

输出速度

共 5 个模型

价格最低

#1
Kimi K2.6Kimi K2.6
$0.70
#2
Kimi K2.6 (non-reasoning)Kimi K2.6 (non-reasoning)
$0.70
#3
Kimi K2.7 CodeKimi K2.7 Code
$0.72
#4
Kimi K3 (max)Kimi K3 (max)
$2.31
#5
Kimi K3 (low)Kimi K3 (low)
$2.31

每 100 万 token 的混合价格

共 5 个模型

Kimi 提供 5 个模型,每个模型的智能、性能和价格特征各不相同。 下方对比了各模型的关键指标。

  • 智能方面,Kimi 上表现最好的模型是 Kimi K3 (max)(44)、Kimi K3 (low)(30)和Kimi K2.6(27)。
  • 输出速度方面,最快的模型是 Kimi K2.7 Code(105 t/s)、Kimi K2.6(71 t/s)和Kimi K2.6 (non-reasoning)(64 t/s)。 各模型之间的速度差异显著,最快与最慢相差 129%。
  • 延迟方面,Kimi K2.6 (non-reasoning)(3.27 秒)、Kimi K2.7 Code(24.52 秒)和Kimi K3 (max)(46.94 秒) 的首个答案 Token 延迟最低。
  • 价格方面,Kimi K2.6($0.70)、Kimi K2.6 (non-reasoning)($0.70)和Kimi K2.7 Code($0.72) 每 100 万 token 的混合价格最低。 各模型价格最多相差 3.3 倍。
  • 上下文窗口方面,Kimi K3 (max)(1M)、Kimi K3 (low)(1M)和Kimi K2.6(256k) 支持 Kimi 上最大的上下文窗口。
Updated
Artificial Analysis Intelligence Index · Higher is better
Output tokens per second · Higher is better
USD per 1M tokens (blended) · Lower is better

智能评测

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.3.2 incorporates 10 evaluations: AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1
Estimate (independent evaluation forthcoming)

Intelligence Evaluations

Intelligence evaluations measured independently by Artificial Analysis · Higher is better
查看更多

Agentic knowledge work, (Elo-500)/2000

Agentic real-world work tasks, (Elo-500)/2000

Agentic SaaS workflows

Agentic coding & terminal use

SciCodeUnder review

Coding

Reasoning & knowledge

Professional document reasoning, All-pass

CritPtUnder review

Physics reasoning

Long context reasoning

Legal agentic work, criterion pass rate

Agentic business operations

Agentic scientific research workflows in a terminal

No data available

Quantitative analysis on spreadsheets & documents

Kubernetes incident root-cause analysis

Visual reasoning

Medical long context reasoning

Intelligence Index 与价格

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended)
Most attractive quadrant
Pareto line

上下文窗口

上下文窗口

Context window: tokens limit · Higher is better

价格

Intelligence Index 与价格

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended)
Most attractive quadrant
Pareto line

性能摘要

输出速度与价格

Output speed: output tokens per second · USD per 1M tokens (blended)
Most attractive quadrant
Pareto line

速度

按输出速度(每秒 token 数)衡量

输出速度

Output tokens per second · Higher is better

延迟

按首 Token 延迟(秒)衡量

延迟: 首个答案 Token 延迟

收到首个回答 token 的秒数 · 包含推理模型的“思考”时间

端到端响应时间

Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed

端到端响应时间与价格

End-to-end response time: end-to-end seconds to output 500 tokens · USD per 1M tokens (blended)
Most attractive quadrant

进一步分析
Kimi 标志
Kimi K3 (max)
1.05M
开放
44
$2.00
34
4.53
78.63
59.28
Kimi 标志
Kimi K3 (low)
1.05M
开放
30
$1.15
36
5.58
74.85
55.41
Kimi 标志
Kimi K2.6
256k
开放
27
$0.80
69
2.97
74.90
64.66
Kimi 标志
Kimi K2.7 Code
256k
开放
26
$0.54
83
2.95
35.88
26.90
Kimi 标志
Kimi K2.6 (non-reasoning)
256k
开放
24*
--
49
2.75
12.89
--

关键定义

常见问题

关于 Kimi 的常见问题