How much does Kimi K2.7 Code cost?

Kimi K2.7 Code costs $0.61 per million input tokens and $3.07 per million output tokens. For a typical conversation (~2,000 tokens), that's approximately $0.004 per message.

What benchmarks has Kimi K2.7 Code been tested on?

Kimi K2.7 Code has been evaluated on 17 benchmarks. Top scores: OTIS Mock AIME 2024-2025: 96.4, GPQA diamond: 86.0, LiveBench — Reasoning: 82.8.

Is Kimi K2.7 Code open source?

Yes, Kimi K2.7 Code is open source.

How does Kimi K2.7 Code compare to Qwen2.5 Coder 32B Instruct?

Kimi K2.7 Code has an average score of 71.7 while Qwen2.5 Coder 32B Instruct scores 71.6. Kimi K2.7 Code outperforms Qwen2.5 Coder 32B Instruct overall. Kimi K2.7 Code costs $0.61/1M input vs Qwen2.5 Coder 32B Instruct at $0.66/1M input. See full comparison →

Home/Models/Kimi K2.7 Code

Kimi K2.7 Code

Name: Kimi K2.7 Code
Price: 0.612 USD
Author: moonshotai

by moonshotai · Released Jun 2026

Open SourceMultimodal

71.7

avg score

Rank #45

Compare

Better than 84% of all models

Context

262K tokens (~131 books)

Input $/1M

$0.61

Output $/1M

$3.07

Type

multimodal

License

Open Source

Benchmarks

17 tested

Data updated today

About

MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts. It uses a native multimodal mixture-of-experts...

Tested on 17 benchmarks with 63.1% average. Top scores: OTIS Mock AIME 2024-2025 (96.4%), GPQA diamond (86.0%), LiveBench — Reasoning (82.8%).

Looking for similar performance at lower cost?
MiniMax M2 scores 72.4 (101% as good) at $0.26/1M input · 58% cheaper

Capabilities

coding

72.0

#19 globally

reasoning

72.7

#31 globally

math

60.6

#61 globally

knowledge

54.5

#91 globally

language

67.1

#87 globally

speed

73.5

#20 globally

Benchmark Scores

Compare All

Tested on 17 benchmarks · Ranked across 6 categories

Score Distribution (all 274 models)

0255075100

▲ You are here

codingCompare coding →

LiveBench — Coding

Regularly refreshed coding problems that avoid data contamination. New problems added monthly to prevent memorization.

74.0—

LiveBench — Agentic Coding

LiveBench coding tasks that require multi-step reasoning and tool use. Tests planning and execution of complex coding workflows.

70.0—

reasoningCompare reasoning →

LiveBench — Reasoning

Regularly refreshed reasoning problems testing logical deduction, spatial reasoning, and analytical thinking.

82.8—

LiveBench — Data Analysis

Fresh data analysis tasks testing ability to interpret tables, charts, and statistical data.

62.7—

mathCompare math →

OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

96.4—

LiveBench — Mathematics

Regularly updated math problems that test numerical reasoning, algebra, calculus, and combinatorics.

79.6—

FrontierMath-Tiers-1-3-v2-Private

54.0—

Quick compare:

vs Qwen2.5 Coder 32B Instruct

vs o4 Mini High

vs Qwen2.5 14B Instruct

Excellent (85+) Good (70-85) Average (50-70) Below (<50)

Similar Models

Qwen2.5 Coder 32B Instruct

Links

Info

moonshotai Pricing explorer Developers · API

Research

Documentation

Community

Source Code

BenchGecko API

kimi-k2-7-code

Specifications

Typemultimodal
Context262K tokens (~131 books)
ReleasedJun 2026
LicenseOpen Source
StatusActive
Cost / Message~$0.004

Available On

moonshotai$0.61

Frequently Asked Questions

Kimi K2.7 Code is an open-source multimodal AI model by moonshotai, released in June 2026. It has an average benchmark score of 71.7. Context window: 262K tokens.

Benchmarks

OTIS Mock AIME 2024-2025 GPQA diamond LiveBench — Reasoning LiveBench — Mathematics LiveBench — Language

moonshotai · Provider moonshotai · Economy All Models Compare Models Pricing Developers · API Context Window · Glossary

Kimi K2.7 Code

Frequently Asked Questions

Related Models

Benchmarks

Related Pages