Llamar a la API de Anthropic: messages

¿Por qué Anthropic?

Install the SDK

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

Su primera llamada

response = client.messages.create(
    model='claude-sonnet-4-5',
    max_tokens=512,
    system='You are a concise assistant.',
    messages=[
        {'role': 'user', 'content': 'Capital of France?'},
    ],
)
print(response.content[0].text)
# 'Paris.'

Diferencias clave con OpenAI

Bloques de contenido

response.content              # list of blocks
response.content[0].type      # 'text' or 'tool_use'
response.content[0].text      # the text content
response.stop_reason          # 'end_turn' / 'max_tokens' / 'tool_use'
response.usage.input_tokens
response.usage.output_tokens

Contenido de varios bloques

messages = [{
    'role': 'user',
    'content': [
        {'type': 'text', 'text': 'What is in this image?'},
        {'type': 'image', 'source': {
            'type': 'base64',
            'media_type': 'image/png',
            'data': image_base64
        }}
    ]
}]

Prellenado del assistant

messages = [
    {'role': 'user', 'content': 'Return a JSON object.'},
    {'role': 'assistant', 'content': '{'}
]
response = client.messages.create(
    model='claude-sonnet-4-5',
    max_tokens=512,
    messages=messages,
)
# Output starts after '{', guaranteed to be JSON.

Uso de herramientas

tools = [{
    'name': 'get_weather',
    'description': 'Get current weather',
    'input_schema': {
        'type': 'object',
        'properties': {'city': {'type': 'string'}},
        'required': ['city']
    }
}]

response = client.messages.create(
    model='claude-sonnet-4-5',
    max_tokens=1024,
    tools=tools,
    messages=messages,
)
if response.stop_reason == 'tool_use':
    for block in response.content:
        if block.type == 'tool_use':
            print(block.name, block.input)

Devolución de resultados de herramientas

messages.append({
    'role': 'user',
    'content': [{
        'type': 'tool_result',
        'tool_use_id': 'toolu_abc',
        'content': json.dumps(weather_data)
    }]
})

Almacenamiento en caché de prompts

system = [
    {'type': 'text', 'text': '...long system prompt...',
     'cache_control': {'type': 'ephemeral'}}
]
for block in system:
    print(f"type={block['type']} cache_control={block['cache_control']}")
    print("text preview:", block['text'][:30])

Razonamiento extendido

response = client.messages.create(
    model='claude-sonnet-4-5',
    max_tokens=4096,
    thinking={'type': 'enabled', 'budget_tokens': 2048},
    messages=messages,
)

Ubicación del prompt del sistema

Repaso