DeepSeek API: Powerful Models for Developers and Teams

Getting Started with the API

Setting up API access begins with creating a developer account at the DeepSeek platform and generating your first API key from the credentials section. The quickstart process involves three core steps: authentication configuration, SDK installation, and executing your initial request. Most developers complete first request testing within 15 minutes using provided code templates.

Authentication uses bearer token format with keys prefixed by "sk-". The base URL for all API endpoints is https://api.deepseek.com/v1, following RESTful conventions. Required headers include Authorization with your API key and Content-Type set to application/json. Rate limiting applies per-key rather than per-account, allowing teams to distribute quotas across multiple projects.

For the Python SDK installation, use pip to add the official client library. The following code demonstrates a complete first request workflow using the chat completion endpoint with DeepSeek V3:

pip install deepseek-sdk

from deepseek import DeepSeek

client = DeepSeek(api_key="sk-your-api-key-here")

response = client.chat.completions.create(
    model="deepseek-chat-v3",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Explain quantum computing in simple terms."}
    ],
    max_tokens=500,
    temperature=0.7
)

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

For developers preferring curl example requests, the equivalent HTTP call requires explicit header configuration. This approach works for testing without SDK dependencies:

curl https://api.deepseek.com/v1/chat/completions \
  -H "Authorization: Bearer sk-your-api-key-here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-chat-v3",
    "messages": [{"role": "user", "content": "Hello, API!"}],
    "max_tokens": 100
  }'

The API returns JSON responses containing generated text, token usage statistics, and request metadata. Successful responses include a choices array with the model's output, while errors return standardized codes for debugging. Token counts appear in the usage object, tracking prompt_tokens, completion_tokens, and total_tokens for billing accuracy.

• Retrieve your API key setup from the developer dashboard security tab.
• Install the Python SDK or use direct HTTP requests for language flexibility.
• Test connectivity with a simple chat completion before production integration.
• Monitor the response headers for rate limit status and remaining quota.

API quickstart guides in the documentation cover additional languages including Node.js and Go, with framework-specific examples for Express, Flask, and FastAPI integrations. Webhook configurations for asynchronous processing require endpoint verification during initial setup.