NOUMENOpen app
NOUMEN

Moonshot · Open weights

Kimi K3 TEE

Added Jul 31, 2026Open weights

Compare models

Choose up to 3 models to compare.0 of 3 chosen

Loading models…

Use in the app

Kimi K3 is Moonshot AI's open-weight, always-thinking multimodal model for long-context reasoning, coding, tool use, and native image understanding. Running inside a TEE (Trusted Execution Environment), with provider attestation support.

  • Reasoning
  • Tools
  • Vision
  • Structured output
Input
Text, Images
Output
Text
Reasoning effort
low · high · max
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$3.00
Output
$15.00
Cache read
$1.50

Reference prices as published by the model provider. API example

Specifications

Context window
1M
Max output
65.5K
Average output (7 days)
1.1K tokens
Longer than 75% of 473 models

Side by side

Compare models

Put Kimi K3 TEE next to up to three other models. The comparison stays in the link, so you can share it.

Compare models

Choose up to 3 models to compare.0 of 3 chosen

Loading models…

TryStep 5 PreviewClaude Haiku 5.5Grok 4.6All 3

Public evaluations

Benchmarks

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

Intelligence Index

43.6

Tested at max reasoning effort

Better than 89% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    76.2

    Composite of the coding evaluations

    Better than 91% of 168 scored models

Agentic work

  • AutomationBench-AA

    58.3%

    Workflow automation, penalized for guardrail violations

    Better than 75% of 173 scored models

  • AutomationBench-AA tasks completed

    24.8%

    Workflows finished end to end without a guardrail violation

    Better than 72% of 173 scored models

  • Harvey LAB-AAv1.1

    5.3%

    Legal tasks passed in full, with no material hallucination

    Better than 82% of 29 scored models

  • AA-Briefcasev1.1

    1,499 Elo

    Agentic knowledge work, rated head to head

    Better than 83% of 166 scored models

  • GDPval-AAv2.1

    1,534 Elo

    Economically valuable occupational tasks, rated head to head

    Better than 80% of 187 scored models

Document reasoning

  • GDP.pdf

    22.0%

    Professional reasoning over PDF documents, all-pass rate

    Better than 80% of 171 scored models

  • AA-LCRv1.1

    88.7%

    Reasoning across long documents

    Better than 99% of 309 scored models

  • MLCR-AA

    21.7%

    Long-context reasoning over medical records

    Better than 84% of 121 scored models

Reasoning and knowledge

  • Humanity’s Last Exam

    46.9%

    Expert-level questions across academic fields

    Better than 90% of 323 scored models

  • GPQA Diamond

    93.5%

    Graduate-level science questions

    Better than 92% of 291 scored models

Coding

  • Terminal-Benchv4.0

    12.6%

    Practical software engineering in a terminal

    Better than 65% of 171 scored models

  • SciCode

    59.5%

    Scientific computing in Python

    Better than 92% of 174 scored models

Legacy evaluations

  • AA-LCR (unversioned)

    88.7%

    Long-context reasoning, earlier grading

    Better than 99% of 309 scored models

Scores from Artificial Analysis, updated Oct 26, 2026. Kimi K3 TEE was tested at max reasoning effort; compare scores measured at the same effort.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
42.1 tok/s
Faster than 28% of 260 models
Time to first token
2.5 s
Quicker than 67% of 260 models
Prompt cache hits
84.5%
About $1.73 per 1M input tokens with caching
Request success (24h)
97.9%
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 61.6%
  • 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": "TEE/kimi-k3",
    "reasoning_effort": "low",
    "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="TEE/kimi-k3",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Moonshot

Other Moonshot models, starting with the same family.