Getting Started with DeepSeek API: Your First API Call

The DeepSeek API provides powerful AI capabilities, and the best part is its compatibility with the OpenAI API format. This means you can leverage existing OpenAI SDKs and tools with minimal configuration changes to access DeepSeek's advanced models. This guide will walk you through making your first API call to the DeepSeek API.

DeepSeek API: OpenAI Compatibility

One of the key advantages of the DeepSeek API is its design to be compatible with the OpenAI API. By adjusting the base URL, you can seamlessly integrate DeepSeek's models into your existing projects that already use the OpenAI SDK or other OpenAI-compatible software.

Setting Up Your Environment

Before you can make your first API call, you need to:

1. Obtain an API Key: You can apply for an API key on the DeepSeek Platform.
2. Choose a Base URL: Use https://api.deepseek.com or https://api.deepseek.com/v1 as your base URL. Note that the /v1 in the URL does not correspond to the model version but is for compatibility.

Making Your First API Call

Here are examples using curl, Python, and Node.js to invoke the Chat API. This example demonstrates a non-streaming request. To receive a stream of responses, set the stream parameter to true.

cURL

curl https://api.deepseek.com/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <DeepSeek API Key>" \
  -d '{
    "model": "deepseek-chat",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Hello!"}
    ],
    "stream": false
  }'

Python

Make sure you have the OpenAI SDK installed:

pip3 install openai

Then, use the following code:

from openai import OpenAI
client = OpenAI(api_key="<DeepSeek API Key>", base_url="https://api.deepseek.com")

response = client.chat.completions.create(
  model="deepseek-chat",
  messages=[
    {"role": "system", "content": "You are a helpful assistant"},
    {"role": "user", "content": "Hello"},
  ],
  stream=False
)

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

Node.js

First, install the OpenAI SDK:

npm install openai

Then, use this code:

import OpenAI from "openai";

const openai = new OpenAI({
  baseURL: 'https://api.deepseek.com',
  apiKey: '<DeepSeek API Key>'
});

async function main() {
  const completion = await openai.chat.completions.create({
    messages: [{ role: "system", content: "You are a helpful assistant." }],
    model: "deepseek-chat",
  });
  console.log(completion.choices[0].message.content);
}

main();

Key Considerations

• Model Selection: The deepseek-chat model has been upgraded to DeepSeek-V3, with the API remaining the same. You can invoke DeepSeek-V3 by specifying model='deepseek-chat'. For reasoning tasks, consider using the reasoning model, DeepSeek-R1, released by DeepSeek. You can invoke DeepSeek-R1 by specifying model='deepseek-reasoner'. See more on Reasoning Models.
• Streaming: For real-time applications, setting stream to true enables you to receive responses as they are generated
• API Reference: for information on the API endpoints visit the API Reference

Further Exploration

Now that you've made your first API call, explore other features and capabilities of the DeepSeek API. Check out the documentation for:

• Models & Pricing: Understand the different models available and their associated costs.
• The Temperature Parameter: Learn how to control the randomness and creativity of the model's responses.
• Token & Token Usage: Get insights into how tokens are used and managed.
• Function Calling: Discover how to enable the model to call external functions.
• Error Codes: Understand common error codes and how to resolve them.

By leveraging the DeepSeek API, you can integrate cutting-edge AI capabilities into your applications with ease.