---
license: mit
tags:
  - tutorial
  - crazyrouter
  - openai
  - api-gateway
  - getting-started
  - llm
language:
  - en
  - zh
---

# 🚀 Crazyrouter Getting Started Guide

> One API key. 624+ AI models. OpenAI-compatible. Get started in 60 seconds.

## What is Crazyrouter?

[Crazyrouter](https://crazyrouter.com/?utm_source=huggingface&utm_medium=tutorial&utm_campaign=dev_community) is an AI API gateway. Instead of managing separate API keys for OpenAI, Anthropic, Google, DeepSeek, and dozens of other providers, you use **one key** to access them all.

It's fully **OpenAI SDK compatible** — just change the `base_url` and you're done.

---

## Step 1: Get Your API Key

1. Go to [crazyrouter.com](https://crazyrouter.com/?utm_source=huggingface&utm_medium=tutorial&utm_campaign=dev_community)
2. Sign up (free tier available)
3. Copy your API key (`sk-...`)

---

## Step 2: Install the OpenAI SDK

```bash
pip install openai
```

That's it. No special SDK needed.

---

## Step 3: Make Your First API Call

### Python

```python
from openai import OpenAI

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

# Use any model — GPT-4o, Claude, Gemini, DeepSeek...
response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[
        {"role": "user", "content": "What is an API gateway?"}
    ]
)

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

### cURL

```bash
curl https://crazyrouter.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-your-crazyrouter-key" \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
```

### JavaScript / Node.js

```javascript
import OpenAI from "openai";

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

const response = await client.chat.completions.create({
  model: "gpt-4o-mini",
  messages: [{ role: "user", content: "Hello from Node.js!" }],
});

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

---

## Step 4: Switch Models Instantly

The magic of Crazyrouter — switch between any model by changing one string:

```python
# OpenAI
response = client.chat.completions.create(model="gpt-4o", messages=messages)

# Anthropic Claude
response = client.chat.completions.create(model="claude-sonnet-4-20250514", messages=messages)

# Google Gemini
response = client.chat.completions.create(model="gemini-2.0-flash", messages=messages)

# DeepSeek
response = client.chat.completions.create(model="deepseek-chat", messages=messages)

# DeepSeek R1 (reasoning)
response = client.chat.completions.create(model="deepseek-reasoner", messages=messages)
```

No SDK changes. No config changes. Just the model name.

---

## Step 5: Streaming Responses

```python
stream = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Write a short poem about coding"}],
    stream=True
)

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

---

## Step 6: List Available Models

```python
models = client.models.list()
for model in models.data:
    print(model.id)
```

Or via cURL:

```bash
curl https://crazyrouter.com/v1/models \
  -H "Authorization: Bearer sk-your-crazyrouter-key"
```

---

## Popular Models

| Model | Provider | Best For |
|-------|----------|----------|
| `gpt-4o` | OpenAI | General purpose, vision |
| `gpt-4o-mini` | OpenAI | Fast, cheap, good enough |
| `claude-sonnet-4-20250514` | Anthropic | Coding, analysis |
| `claude-haiku-3.5` | Anthropic | Fast, affordable |
| `gemini-2.0-flash` | Google | Speed, long context |
| `deepseek-chat` | DeepSeek | Cheap, strong coding |
| `deepseek-reasoner` | DeepSeek | Complex reasoning (R1) |
| `o3-mini` | OpenAI | Reasoning tasks |

---

## Environment Variables (Recommended)

```bash
export OPENAI_API_KEY="sk-your-crazyrouter-key"
export OPENAI_BASE_URL="https://crazyrouter.com/v1"
```

Then your code simplifies to:

```python
from openai import OpenAI
client = OpenAI()  # Reads from env vars automatically
```

---

## FAQ

**Q: Is it really OpenAI-compatible?**
A: Yes. Any tool that works with the OpenAI API works with Crazyrouter. Just change the base URL.

**Q: Is there a free tier?**
A: Yes. Sign up at [crazyrouter.com](https://crazyrouter.com/?utm_source=huggingface&utm_medium=tutorial&utm_campaign=dev_community) to get started.

**Q: How many models are available?**
A: 624+ models from 20+ providers, and growing.

**Q: Can I use it with LangChain / LlamaIndex / AutoGen?**
A: Absolutely. See our [LangChain guide](https://huggingface.co/xujfcn/Crazyrouter-LangChain-Guide).

---

## Links

- 🌐 [Website](https://crazyrouter.com/?utm_source=huggingface&utm_medium=tutorial&utm_campaign=dev_community)
- 🤖 [Live Demo](https://huggingface.co/spaces/xujfcn/Crazyrouter-Demo)
- 💰 [Pricing](https://huggingface.co/spaces/xujfcn/Crazyrouter-Pricing)
- 📊 [Model Pricing Dataset](https://huggingface.co/datasets/xujfcn/Crazyrouter-Model-Pricing)
- 💬 [Telegram Community](https://t.me/crazyrouter)
- 🐦 [Twitter @metaviiii](https://twitter.com/metaviiii)