Kimi: Models Intelligence, Performance & Price

Kimi
Kimi

This analysis is intended to support you in choosing the best model provided by Kimi for your use-case.

Most Intelligent

#1
Kimi K3.1 ProKimi K3.1 Pro
46
#2
Kimi K3.1 Pro High SpeedKimi K3.1 Pro High Speed
46
#3
Kimi K3 (low)Kimi K3 (low)
17

Intelligence Index

Total 7 models

Lowest Time per Task

#1
Kimi K3.1 Pro High SpeedKimi K3.1 Pro High Speed
16.3 min
#2
Kimi K3 (low)Kimi K3 (low)
16.8 min
#3
Kimi K3.1 ProKimi K3.1 Pro
38.6 min

Time per Intelligence Index task (minutes)

Total 7 models

Lowest Price

#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 K3.1 ProKimi K3.1 Pro
$1.54
#5
Kimi K3 (low)Kimi K3 (low)
$2.31

Blended price (per 1M tokens)

Total 7 models

Kimi offers 7 models, each with different intelligence, performance, and pricing characteristics. Below is a comparison of the key metrics across models.

  • For intelligence, the top models on Kimi are Kimi K3.1 Pro (46), Kimi K3.1 Pro High Speed (46), and Kimi K3 (low) (17).
  • For time per task, Kimi K3.1 Pro High Speed (16.3 min), Kimi K3 (low) (16.8 min), and Kimi K3.1 Pro (38.6 min) finish an Intelligence Index task fastest on Kimi.
  • For latency, Kimi K3 (low) (6.71s), Kimi K2.6 (non-reasoning) (7.14s), and Kimi K2.7 Code (36.20s) offer the lowest time to first answer token.
  • For pricing, Kimi K2.6 ($0.70), Kimi K2.6 (non-reasoning) ($0.70), and Kimi K2.7 Code ($0.72) offer the lowest blended prices per 1M tokens. Prices vary up to 3.3x across models.
  • For context window size, Kimi K3 (low) (1M), Kimi K3 (max) (1M), and Kimi K3.1 Pro (1M) support the largest context windows on Kimi.
Artificial Analysis Intelligence Index · Higher is better
Inference time (minutes) per Intelligence Index task · Lower is better
USD per 1M tokens (blended) · Lower is better

Intelligence

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v5.0.1 incorporates 9 evaluations: AA-Briefcase v1.1, AutomationBench-AA, Terminal-Bench 4.0, FrontierCode-AA 2.0, AA-Omniscience v1.1, GDP.pdf, HLE-Diamond, CritPt v2 (Hard Subset), and Terminal-Bench-Science 0.1

Evaluations

Evaluations measured independently by Artificial Analysis · AA-PersonalityBench scores are percentile ranks from 1 to 100
See more

Agentic knowledge work, Elo

Agentic SaaS workflows

Agentic scientific research workflows in a terminal

Agentic software engineering in open-source repositories

Reasoning & knowledge

Professional document reasoning, All-pass

Agentic coding & terminal use

Physics reasoning

Quantitative analysis on spreadsheets & documents

Legal agentic work, Hallucination-Gated All-Pass Rate

Agentic business operations

Agentic real-world work tasks, Elo

Agentic interactive 3D experiences, Elo

Product and brand videos designed in code, Elo

Kubernetes incident root-cause analysis

Medical long context reasoning

Visual reasoning

Rich, varied, well-crafted language

Well-structured, scannable, answer-first

Truthful, acknowledges uncertainty, non-sycophantic

Use of filler and generic language

Intelligence Index vs. Price

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

Pricing

Cost per Intelligence Index Task

Weighted average cost (USD) per Intelligence Index task · Segmented by token type · Lower is better

Intelligence Index vs. Price

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

Speed

Average of the single-request and multi-turn agentic workloads measured

Time per Intelligence Index Task

Output tokens divided by end-to-end output speed (minutes per Intelligence Index task) · Lower is better

Quality

Endpoint Accuracy Index

Each endpoint's accuracy as a % of its model's reference endpoint · Color shows whether the gap is statistically significant

Latency

Average of the single-request and multi-turn agentic workloads measured

Time to First Answer Token

Seconds until the first answer token after any reasoning · Average of the single-request and multi-turn agentic workloads measured · Lower is better

Reliability

Share of requests the endpoint failed, across all the single-request and multi-turn agentic workloads measured

Endpoint Error Rate

Share of requests the endpoint failed with an error, a rate limit or a timeout · Across all the single-request and multi-turn agentic workloads measured · Lower is better

Caching

Cache Hit Rate

Median across multi-turn agentic session runs in the last four weeks, updated Oct 26, 2026 · Higher is better

Cost per Task vs. Cache Hit Rate

Weighted average cost (USD) per Intelligence Index task · Cache hit rate: median of the last four weeks, updated Oct 26, 2026
Most attractive quadrant
Pareto line

Further Analysis
Kimi logo
Kimi K3.1 Pro
46
$3.68
38.6
99%
153.79
5.4%
1M
Open
Kimi logo
Kimi K3.1 Pro High Speed
46
$4.42
16.3
99%
71.18
5.5%
1M
Open
Kimi logo
Kimi K3 (low)
17
$2.37
16.8
n/a
6.71
0.0%
1M
Open
Kimi logo
Kimi K2.6
--
--
--
n/a
89.80
0.0%
256k
Open
Kimi logo
Kimi K2.6 (non-reasoning)
--
--
--
n/a
7.14
0.0%
256k
Open
Kimi logo
Kimi K2.7 Code
--
--
--
n/a
36.20
0.0%
256k
Open
Kimi logo
Kimi K3 (max)
--
--
--
100%
36.41
0.0%
1M
Open

Key definitions

Frequently Asked Questions

Common questions about Kimi