Kimi K2.5 API — Complete Reference for Developers 2026

What is Kimi K2.5 API

The kimi k2.5 api is the latest flagship model from Moonshot AI, released in 2025 and continuously updated through 2026. It represents the most capable model in the Kimi family, excelling at complex reasoning, code generation, mathematical problem-solving, and multimodal understanding (vision inputs).

The kimi-k2.5 api follows the OpenAI chat completions format, meaning developers familiar with GPT-4 can integrate Kimi K2.5 with minimal code changes. The model is identified as kimi-k2-5 in API requests, with a thinking variant available as kimi-k2-5-thinking.

Kimi K2.5 Specifications

How to Call Kimi K2.5 API with Python and cURL

Here's how to make a basic kimi k2.5 api call using Python with the OpenAI SDK:

from openai import OpenAI

client = OpenAI(
    api_key="your-kimi-api-key",
    base_url="https://api.moonshot.cn/v1"
)

response = client.chat.completions.create(
    model="kimi-k2-5",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Explain quantum computing in simple terms."}
    ],
    temperature=0.7,
    max_tokens=2048
)

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

curl https://api.moonshot.cn/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer your-kimi-api-key" \
  -d '{
    "model": "kimi-k2-5",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Explain quantum computing in simple terms."}
    ],
    "temperature": 0.7,
    "max_tokens": 2048
  }'

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'your-kimi-api-key',
  baseURL: 'https://api.moonshot.cn/v1'
});

const response = await client.chat.completions.create({
  model: 'kimi-k2-5',
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'Explain quantum computing in simple terms.' }
  ],
  temperature: 0.7,
  max_tokens: 2048
});

console.log(response.choices[0].message.content);

Is Kimi K2.5 API Better Than GPT-4o

Kimi K2.5 API is approximately 44% cheaper than GPT-4o for input tokens and 58% cheaper for output tokens, while achieving competitive benchmark scores. For developers prioritizing cost efficiency without sacrificing quality, kimi k2.5 api is a strong alternative to OpenAI's offerings.

Kimi K2.5 API Pricing

For detailed kimi k2.5 api pricing including volume discounts, enterprise plans, and free tier limits, see our complete Kimi API Pricing guide. The kimi 2.5 api cost is among the most competitive for a model of this capability level.

Summary

The kimi k2.5 api delivers flagship-level AI capabilities at a fraction of GPT-4o's cost, with free access available through Nvidia's API Catalog. Start with our API Key guide to get your first K2.5 call running in minutes.