NOUMENOpen app
NOUMEN

Moonshot · Open weights

Kimi K2.7 Code

Added Jun 12, 2026Open weights

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Use in the app

Kimi K2.7 Code is Moonshot AI's coding-focused agentic model built for long-horizon software engineering workflows. It supports native image input, tool calling, and forced thinking mode; instant/non-thinking mode is not supported.

  • Reasoning
  • Tools
  • Vision
  • Parallel tools
Input
Text, Images
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.67
Output
$3.35
Cache read
$0.15

Reference prices as published by the model provider. API example

Specifications

Context window
262.1K
Max output
65.5K
Average output (7 days)
1.8K tokens
Longer than 89% of 473 models

Side by side

Kimi K2.7 Code compared

Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Kimi K2.7 CodeMoonshotThis modelKimi K3.1 Pro HighspeedMoonshotRemove
ProviderMoonshotMoonshot
Discovery tierOpen weightsFrontier
AddedJun 12, 2026Oct 24, 2026
Limits
Context window262.14K1.05MBest
Max output65.54K—
Average output1.8K tokens1.3K tokens
Pricing, USD per 1M tokens
Input$0.67Best$4.00
Output$3.35Best$20.00
Cache read$0.15Best$0.20
Cache write—$4.00
Performance
Throughput72.0 tok/s—
Time to first token3.0 s—
Request success (24h)99.7%99.8%Best
Benchmark indexes
Intelligence Index25.855.1Best
Coding Index60.8—
Capabilities
ReasoningYesYes
ToolsYesYes
VisionYesYes
Structured outputNoYes
Video inputNoYes
Parallel toolsYesNo
Reasoning effort—low, high, max

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

  • Kimi K2.7 Code
  • Kimi K3.1 Pro Highspeed

Intelligence Index

  • Kimi K2.7 Code25.8
  • Kimi K3.1 Pro Highspeed55.1

Kimi K2.7 Code: Better than 63% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • Kimi K2.7 Code: 60.8
    • Kimi K3.1 Pro Highspeed: —

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Kimi K2.7 Code: 24.5%
    • Kimi K3.1 Pro Highspeed: 73.3%Top

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Kimi K2.7 Code: 4.7%
    • Kimi K3.1 Pro Highspeed: 46.0%Top

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Kimi K2.7 Code: 858 Elo
    • Kimi K3.1 Pro Highspeed: 1,678 EloTop

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Kimi K2.7 Code: 1,038 Elo
    • Kimi K3.1 Pro Highspeed: 1,846 EloTop

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Kimi K2.7 Code: 11.2%
    • Kimi K3.1 Pro Highspeed: 31.4%Top

  • AA-LCRv1.1

    Reasoning across long documents

    • Kimi K2.7 Code: 79.3%
    • Kimi K3.1 Pro Highspeed: 86.3%Top

  • MLCR-AA

    Long-context reasoning over medical records

    • Kimi K2.7 Code: 12.2%
    • Kimi K3.1 Pro Highspeed: —

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Kimi K2.7 Code: 35.0%
    • Kimi K3.1 Pro Highspeed: 52.5%Top

  • GPQA Diamond

    Graduate-level science questions

    • Kimi K2.7 Code: 89.6%
    • Kimi K3.1 Pro Highspeed: —

  • IFBench

    Following precise output instructions

    • Kimi K2.7 Code: 63.1%
    • Kimi K3.1 Pro Highspeed: —

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Kimi K2.7 Code: 1.0%
    • Kimi K3.1 Pro Highspeed: 67.2%Top

  • SciCode

    Scientific computing in Python

    • Kimi K2.7 Code: 47.8%
    • Kimi K3.1 Pro Highspeed: 60.8%Top

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • Kimi K2.7 Code: 44.7%
    • Kimi K3.1 Pro Highspeed: —

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Kimi K2.7 Code: 90.1%
    • Kimi K3.1 Pro Highspeed: —

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Kimi K2.7 Code: 79.3%
    • Kimi K3.1 Pro Highspeed: 86.3%Top

Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis

Live routing data

Performance and reliability

Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.

Throughput
72.0 tok/s
Faster than 49% of 256 models
Time to first token
3.0 s
Quicker than 50% of 256 models
Prompt cache hits
68.3%
About $0.32 per 1M input tokens with caching
Request success (24h)
99.7%
Reliable

Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.

Hourly request success · last 72 hoursLowest hour 98.7%
  • Healthy (98% or more)
  • Degraded (90–98%)
  • Down (under 90%)
  • No requests

Developers

Use it through the API

Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.

Read the quickstart
curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "moonshotai/kimi-k2.7-code",
    "messages": [{ "role": "user", "content": "Explain this stack trace." }]
  }'
import os
from openai import OpenAI

client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
    model="moonshotai/kimi-k2.7-code",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Moonshot

Other Moonshot models, starting with the same family.