Kimi K2

Chinese ultra-long context model supporting 2 million characters input, excelling at long document analysis and processing

Category Text Model ID moonshotai/kimi-k2-instruct-0905

Context Length2M Chinese characters

Model TypeLarge Language Model (LLM)

Special CapabilitiesUltra-long context processing

Pricing & Specs

💰 Pricing

Input¥1 / M tokens

Output¥3 / M tokens

⚙️ Specs

Context Length2M Chinese characters

Model TypeLarge Language Model (LLM)

Special CapabilitiesUltra-long context processing

Language OptimizationChinese

API Examples

Python

from openai import OpenAI

client = OpenAI(
    base_url="https://api.zairouter.com/v1",
    api_key="your-api-key"
)

response = client.chat.completions.create(
    model="moonshotai/kimi-k2-instruct-0905",
    messages=[
        {"role": "user", "content": "Hello!"}
    ]
)

print(response.choices[0].message.content)

cURL

curl https://api.zairouter.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "moonshotai/kimi-k2-instruct-0905",
    "messages": [
      {"role": "user", "content": "Hello!"}
    ]
  }'

Model Features

Kimi K2 is Moonshot AI's second-generation ultra-long context model, uniquely positioned in long document processing:

Ultra-long Context: Supports 2 million Chinese characters, approximately equivalent to 2-3 full-length novels
Chinese Optimization: Deeply optimized for Chinese language
High Cost-effectiveness: Affordable pricing suitable for large-scale applications
Stable and Reliable: Maintains stable performance in long document scenarios

Use Cases

Long paper and report analysis
Legal contract review
Novel and book content analysis
Large-scale codebase understanding
Multi-document comprehensive analysis

Performance Advantages

Kimi K2 ranks among the world's best in ultra-long context processing capabilities, with particularly outstanding performance in Chinese scenarios, making it the ideal choice for long document processing.

← Back to Models