KI

Kimi K2

Chat modelKimi K2Kimi K2

Kimi-K2-0711-Preview

Release date: 2025-07-11Updated: 2026-06-15 07:18:23.8832,736
Parameters
1T
Context length
131K
Chinese support
Supported
Reasoning ability

Kimi-K2-0711-Preview is an AI model published by Moonshot AI, released on 2025-07-11, for Chat model, with 1T parameters, and 131K context length, requiring about 1.01TB storage, under the Modified MIT license, with a 97.40 score on MATH-500.

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

Kimi K2

Model basics

Reasoning traces
Not supported
Thinking modes
Thinking modes not supported
Context length
131K tokens
Max output length
131K tokens
Model type
Chat model
Modality (in / out)
Text → Text
Release date
2025-07-11
Model file size
1.01TB
MoE architecture
Yes
Total params / Active params
1T / 32B
Knowledge cutoff
No data
Kimi K2

Open source & experience

Kimi K2

Official resources

Kimi K2

API details

API speed
2/5
No public API pricing yet.
Kimi K2

Benchmark Results

Kimi K2 currently shows benchmark results led by Creative Writing (1 / 23, score 88.10), MMLU (12 / 65, score 89.50), MATH-500 (11 / 44, score 97.40). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.

Thinking

General Knowledge

6 evaluations
Benchmark / mode
Score
Rank/total
89.50
12 / 65
81.10
53 / 126
75.10
94 / 179
LiveBench
Standard Mode
48.10
100 / 115
13.30
57 / 65
4.70
156 / 159

Common Sense

1 evaluations
Benchmark / mode
Score
Rank/total
31
21 / 45

Coding and Software Engineer

2 evaluations
Benchmark / mode
Score
Rank/total
53.70
86 / 120
51.80
88 / 108

Math and Reasoning

7 evaluations
Benchmark / mode
Score
Rank/total
97.40
11 / 44
69.60
39 / 62
54
85 / 106
2.10
47 / 60
2
8 / 9
0.50
10 / 10

Writing and Creative Capabilities

1 evaluations
Benchmark / mode
Score
Rank/total
88.10
1 / 23

AI Agent - Tool Usage

1 evaluations
Benchmark / mode
Score
Rank/total
37.50
15 / 35

常识推理

1 evaluations
Benchmark / mode
Score
Rank/total
Simple Bench
Standard Mode
26.30
53 / 63

Agent Level Benchmark

4 evaluations
Benchmark / mode
Score
Rank/total
64.30
27 / 40
64.30
27 / 40
Aider-Polyglot
Standard Mode
59.10
24 / 59

Compare with other models

No curated comparisons for this model yet.

Want a custom combination? Open the compare tool

Kimi K2

Model variants & downloads

Variant nameVersion typeQuantizationModel sizeHuggingFace link
Kimi-K2-Instructℹ️InstructBF161031.15 GBDownload link
Kimi-K2-Baseℹ️BaseBF161031.15 GBDownload link
Kimi K2

Publisher

Kimi-K2-0711-Preview

Model Overview

Kimi-K2-0711-Preview is an AI model published by Moonshot AI, released on 2025-07-11, for Chat model, with 1T parameters, and 131K context length, requiring about 1.01TB storage, under the Modified MIT license, with a 97.40 score on MATH-500.

DataLearner on WeChat

Follow DataLearner on WeChat for AI model updates and research notes.

DataLearner WeChat QR code