Bahasa Melayu

Panduan Pantas CrazyRouter API

Artikel ini memperkenalkan cara untuk menyambung ke Crazyrouter API dengan pantas, membuat panggilan pertama dalam masa 5 minit, menyokong OpenAI, Claude

Crazyrouter Team

January 19, 2026 / 291 views

Crazyrouter

Read the docs Check live pricing Open image tool Create account

Selamat datang ke Crazyrouter! Panduan ini akan membantu anda melengkapkan penyambungan API dalam masa 5 minit dan mula menggunakan model AI terkemuka dunia.

Mengapa memilih Crazyrouter?#

Antara muka bersatu: Satu API Key untuk mengakses 300+ model AI
Serasi dengan OpenAI: Tidak perlu ubah kod, hanya gantikan base_url sahaja
Harga lebih rendah: Lebih murah berbanding harga rasmi, caj mengikut penggunaan tanpa langganan
Kebolehsediaan tinggi: Penghalaan pintar, pindahan ralat automatik

1. Dapatkan API Key#

Lawati laman rasmi Crazyrouter
Daftar akaun dan log masuk ke konsol
Masuk ke halaman "Pengurusan Token"
Klik "Cipta Token"
Salin API Key yang dijana (bermula dengan sk-)

Petua: Sila simpan API Key anda dengan baik, jangan dedahkannya di tempat awam.

2. Hantar permintaan pertama#

Guna curl#

bash

curl https://crazyrouter.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-your-api-key" \
  -H "User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36" \
  -d '{
    "model": "gpt-5.2",
    "messages": [{"role": "user", "content": "Hello, how are you?"}],
    "temperature": 0.7
  }'

Guna Python (disyorkan)#

python

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-api-key",
    base_url="https://crazyrouter.com/v1"
,
    default_headers={
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36"
    }
)

response = client.chat.completions.create(
    model="gpt-5.2",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello, how are you?"}
    ],
    temperature=0.7,
    max_tokens=1000
)

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

Guna Node.js#

javascript

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'sk-your-api-key',
  baseURL: 'https://crazyrouter.com/v1'
});

async function main() {
  const response = await client.chat.completions.create({
    model: 'gpt-5.2',
    messages: [
      { role: 'system', content: 'You are a helpful assistant.' },
      { role: 'user', content: 'Hello，Please introduce yourself' }
    ],
    temperature: 0.7
  });

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

main();

Output beraliran (Streaming)#

python

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-api-key",
    base_url="https://crazyrouter.com/v1"
,
    default_headers={
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36"
    }
)

stream = client.chat.completions.create(
    model="gpt-5.2",
    messages=[{"role": "user", "content": "Write a poem about spring"}],
    stream=True
)

for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

3. Model yang disokong#

Crazyrouter menyokong 300+ model AI, berikut sebahagian model popular:

厂商	模型示例	特点
OpenAI	gpt-5.2, gpt-5, gpt-4o, o3-pro, o3-mini	综合能力强，支持多模态
Anthropic	claude-opus-4.5, claude-4, claude-sonnet-4	长上下文，代码能力强
Google	gemini-3.0-pro, gemini-3-flash, gemini-2.5-pro	多模态，推理能力强
xAI	grok-3, grok-3-mini, grok-3-vision	实时信息，深度搜索
DeepSeek	deepseek-r1, deepseek-v3	性价比高，中文优秀

Lihat senarai model penuh: Model Plaza

4. Penjelasan parameter biasa#

参数	类型	说明
`model`	string	Nama model, contohnya `gpt-5.2`
`messages`	array	Senarai mesej perbualan
`temperature`	number	Tahap kerandoman, 0-2, lalai 1
`max_tokens`	integer	Bilangan token output maksimum
`stream`	boolean	Sama ada output beraliran atau tidak
`top_p`	number	Parameter pensampelan nucleus, 0-1

5. Pengendalian ralat#

python

from openai import OpenAI, APIError, RateLimitError

client = OpenAI(
    api_key="sk-your-api-key",
    base_url="https://crazyrouter.com/v1"
,
    default_headers={
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36"
    }
)

try:
    response = client.chat.completions.create(
        model="gpt-5.2",
        messages=[{"role": "user", "content": "Hello!"}]
    )
except RateLimitError:
    print("请求过于频繁，请稍后重试")
except APIError as e:
    print(f"API 错误: {e}")