How to Use DeepSeek API with Python

This tutorial shows you how to use DeepSeek-v3.2 API with Python, from basic calls to production-ready patterns. All examples use the OpenAI SDK, so if you've used GPT-4 before, you already know 90% of this.

Installation

pip install openai

Basic Chat Completion

from openai import OpenAI

client = OpenAI(
    api_key="nvai-your-api-key",
    base_url="https://aiapi-pro.com/v1"
)

response = client.chat.completions.create(
    model="deepseek-v3.2",
    messages=[
        {"role": "system", "content": "You are a helpful coding assistant."},
        {"role": "user", "content": "Write a Python function to merge two sorted lists"}
    ],
    temperature=0.7,
    max_tokens=2048
)

print(response.choices[0].message.content)
print(f"Tokens used: {response.usage.total_tokens}")

Streaming Responses

Streaming is essential for chat applications. It shows output token-by-token instead of waiting for the full response:

stream = client.chat.completions.create(
    model="deepseek-v3.2",
    messages=[{"role": "user", "content": "Explain async/await in Python"}],
    stream=True
)

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

Async Usage (for Web Apps)

If you're building a FastAPI or async web app, use the async client:

from openai import AsyncOpenAI
import asyncio

client = AsyncOpenAI(
    api_key="nvai-your-api-key",
    base_url="https://aiapi-pro.com/v1"
)

async def ask_deepseek(question: str) -> str:
    response = await client.chat.completions.create(
        model="deepseek-v3.2",
        messages=[{"role": "user", "content": question}]
    )
    return response.choices[0].message.content

# Run multiple requests concurrently
async def main():
    tasks = [
        ask_deepseek("What is a binary tree?"),
        ask_deepseek("What is a hash map?"),
        ask_deepseek("What is dynamic programming?"),
    ]
    results = await asyncio.gather(*tasks)
    for r in results:
        print(r[:100], "...\n")

asyncio.run(main())

Error Handling for Production

from openai import OpenAI, APIError, RateLimitError, APIConnectionError
import time

client = OpenAI(api_key="nvai-your-key", base_url="https://aiapi-pro.com/v1")

def safe_completion(messages, retries=3):
    for attempt in range(retries):
        try:
            return client.chat.completions.create(
                model="deepseek-v3.2",
                messages=messages,
                timeout=30
            )
        except RateLimitError:
            time.sleep(2 ** attempt)  # Exponential backoff
        except APIConnectionError:
            time.sleep(1)
        except APIError as e:
            print(f"API error: {e}")
            break
    return None

Multi-Model Routing

Use different models for different tasks through the same client:

# Same client, different models based on task
def code_review(code):
    return client.chat.completions.create(
        model="deepseek-v3.2",  # Best for code — $0.20/1M
        messages=[{"role": "user", "content": f"Review this code:\n{code}"}]
    )

def quick_classify(text):
    return client.chat.completions.create(
        model="qwen-turbo",  # Cheapest — $0.06/1M
        messages=[{"role": "user", "content": f"Classify: {text}"}]
    )

def free_test(prompt):
    return client.chat.completions.create(
        model="glm-4.6v-flash",  # Completely FREE
        messages=[{"role": "user", "content": prompt}]
    )

Available Models & Pricing

# List all available models
models = client.models.list()
for m in models.data:
    print(f"{m.id:25s} by {m.owned_by}")