KI

Kimi K2.7 Code

Coding modelKimi K2

Kimi-K2.7-Code

Release date: 2026-06-12Updated: 2026-06-14 11:59:28.1791,483
Parameters
1T
Context length
256K
Chinese support
Supported
Reasoning ability

Kimi K2.7 Code 是 Moonshot AI 于 2026 年 6 月 12 日发布的 K2 系列编程专属模型,采用 1T 总参数 / 32B 激活 MoE 架构(384 专家),支持 256K 上下文及文本、图像、视频输入。较 K2.6 在 Kimi Code Bench v2 提升 21.8%(62.0),MCP Mark Verified 达 81.1 超越 Opus 4.8,thinking token 使用量降低约 30%。权重以 Modified MIT 许可开源,API 定价 $0.95/$4.00 每百万 input/output tokens。

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

Kimi K2.7 Code

Model basics

Reasoning traces
Supported
Thinking modes
Thinking Mode (Default)
Context length
256K tokens
Max output length
No data
Model type
Coding model
Modality (in / out)
Text, Image, Video → Text
Release date
2026-06-12
Model file size
595GB
MoE architecture
Yes
Total params / Active params
1T / 32B
Knowledge cutoff
No data
Kimi K2.7 Code

Open source & experience

Kimi K2.7 Code

Official resources

DataLearnerAI blog
No blog post yet
Kimi K2.7 Code

API details

API speed
3/5
💡Default unit: $/1M tokens. If vendors use other units, follow their published pricing.
Standard
TypeConditionInputOutput
Text-$0.950/ 1M$4.00/ 1M
Cache PricingPrompt Cache
TypeTTLWriteRead
Text--$0.190/ 1M
Kimi K2.7 Code

Benchmark Results

Kimi K2.7 Code currently shows benchmark results led by LiveBench (30 / 115, score 71.89), TerminalBench 2.1 (10 / 14, score 67.04), DeepSWE (7 / 9, score 31). 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
Tool usage

General Knowledge

1 evaluations
Benchmark / mode
Score
Rank/total
LiveBench
Standard Mode
71.89
30 / 115

AI Agent - Tool Usage

1 evaluations
Benchmark / mode
Score
Rank/total
TerminalBench 2.1
Thinking ModeTools
67.04
10 / 14

Coding and Software Engineer

1 evaluations
Benchmark / mode
Score
Rank/total
DeepSWE
Standard ModeTools
31
7 / 9

Compare with other models

Kimi K2.7 Code

Publisher

Kimi-K2.7-Code

Model Overview

Kimi K2.7 Code 是 Moonshot AI 于 2026 年 6 月 12 日发布的 K2 系列编程专属模型,采用 1T 总参数 / 32B 激活 MoE 架构(384 专家),支持 256K 上下文及文本、图像、视频输入。较 K2.6 在 Kimi Code Bench v2 提升 21.8%(62.0),MCP Mark Verified 达 81.1 超越 Opus 4.8,thinking token 使用量降低约 30%。权重以 Modified MIT 许可开源,API 定价 $0.95/$4.00 每百万 input/output tokens。

DataLearner on WeChat

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

DataLearner WeChat QR code