How to Use Claude API with Python: Complete Beginner’s Guide (2026)

by

The Anthropic Python SDK makes it simple to integrate Claude into your applications. In this guide you’ll go from zero to a working chatbot in under 10 minutes — covering installation, your first API call, streaming, multi-turn conversations, and error handling.

Prerequisites

Step 1: Install the SDK

pip install anthropic

That’s the only dependency you need. The SDK includes everything: the client, streaming support, and type hints.

Step 2: Set Your API Key

Store your key as an environment variable — never hardcode it in your source files:

export ANTHROPIC_API_KEY="sk-ant-..."

Or create a .env file in your project root:

ANTHROPIC_API_KEY=sk-ant-...

Load it with python-dotenv:

pip install python-dotenv

Step 3: Your First API Call

Create a file main.py and add:

import anthropic

client = anthropic.Anthropic()  # reads ANTHROPIC_API_KEY from env

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Explain what an API is in 2 sentences."}
    ]
)

print(message.content[0].text)

Run it:

python main.py

You’ll get a clean, concise response from Claude. The message.content[0].text contains the text output.

Step 4: Add a System Prompt

A system prompt sets the context and personality for Claude — it’s the first thing Claude reads before any user message:

client = anthropic.Anthropic()

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    system="You are a senior Python developer. Answer concisely with code examples.",
    messages=[
        {"role": "user", "content": "How do I read a JSON file in Python?"}
    ]
)

print(message.content[0].text)

Step 5: Streaming Responses

For a better user experience — especially with long outputs — use streaming so text appears word by word:

import anthropic

client = anthropic.Anthropic()

with client.messages.stream(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Write a Python function to parse CSV files"}]
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)

print()  # newline at the end

Step 6: Multi-Turn Conversations

Build a simple chatbot by keeping track of the message history:

import anthropic

client = anthropic.Anthropic()
conversation_history = []

def chat(user_message: str) -> str:
    conversation_history.append({
        "role": "user",
        "content": user_message
    })

    response = client.messages.create(
        model="claude-sonnet-4-6",
        max_tokens=1024,
        system="You are a helpful AI assistant.",
        messages=conversation_history
    )

    assistant_message = response.content[0].text
    conversation_history.append({
        "role": "assistant",
        "content": assistant_message
    })

    return assistant_message

# Example conversation
print(chat("What is Python?"))
print(chat("What are its main use cases?"))
print(chat("Which one is best for AI development?"))

Each call passes the full history so Claude remembers what was said earlier in the conversation.

Step 7: Error Handling

Always wrap API calls in try/except for production code:

import anthropic

client = anthropic.Anthropic()

try:
    message = client.messages.create(
        model="claude-sonnet-4-6",
        max_tokens=1024,
        messages=[{"role": "user", "content": "Hello!"}]
    )
    print(message.content[0].text)

except anthropic.APIConnectionError as e:
    print(f"Connection error: {e}")
except anthropic.RateLimitError as e:
    print(f"Rate limit hit — slow down: {e}")
except anthropic.APIStatusError as e:
    print(f"API error {e.status_code}: {e.message}")

Available Models

Choose the right Claude model for your use case:

  • claude-opus-4-7 — most capable, best for complex reasoning and analysis
  • claude-sonnet-4-6 — best balance of speed and intelligence (recommended for most apps)
  • claude-haiku-4-5-20251001 — fastest and most affordable, great for simple tasks

Key Parameters

The most important parameters in messages.create():

message = client.messages.create(
    model="claude-sonnet-4-6",   # which Claude model to use
    max_tokens=1024,              # maximum tokens in the response
    temperature=0.7,             # 0 = deterministic, 1 = creative
    system="...",                # system prompt (optional)
    messages=[...]               # conversation history
)

💡 Tip: Use temperature=0 for code generation and factual tasks. Use higher values (0.7–1.0) for creative writing.

Complete Example: Simple CLI Chatbot

import anthropic

def main():
    client = anthropic.Anthropic()
    history = []
    print("Claude Chatbot — type 'quit' to exit\n")

    while True:
        user_input = input("You: ").strip()
        if user_input.lower() in ("quit", "exit"):
            break
        if not user_input:
            continue

        history.append({"role": "user", "content": user_input})

        with client.messages.stream(
            model="claude-sonnet-4-6",
            max_tokens=2048,
            system="You are a helpful assistant.",
            messages=history
        ) as stream:
            print("Claude: ", end="", flush=True)
            response_text = ""
            for text in stream.text_stream:
                print(text, end="", flush=True)
                response_text += text
            print()

        history.append({"role": "assistant", "content": response_text})

if __name__ == "__main__":
    main()

What’s Next?

Now that you have the basics working, here’s what to explore next:

  • Tool use — let Claude call functions and APIs in your code
  • Vision — send images to Claude for analysis
  • Prompt caching — reduce costs on repeated context (up to 90% savings)
  • Batch API — process thousands of requests asynchronously at 50% discount

💡 Resources: Anthropic Docs | Python SDK on GitHub